Risk of severe illness of COVID‐19 patients with NAFLD and increased NAFLD fibrosis scores

Abstract Background There is still little knowledge about the association of liver fibrosis with the clinical outcomes of COVID‐19 patients with non‐alcoholic fatty liver disease (NAFLD). The aim of the study was to determine the association of NAFLD fibrosis score (NFS)–determined liver fibrosis with clinical outcomes of COVID‐19 patients with NAFLD. Methods The NAFLD was diagnosed by the Hepatic Steatosis Index (HSI) in the absence of other causes of chronic liver diseases. NFS was used to evaluate the severity of liver fibrosis. Results A total of 86 COVID‐19 patients with NAFLD were included. The median age was 43.5 years, and 58.1% of patients were male. Thirty‐eight (44.2%) patients had advanced liver fibrosis according to the NFS. Multivariate analysis indicated that concurrent diabetes (odds ratio [OR] 8.264, 95% confidence interval [CI] 1.202–56.830, p = 0.032) and advanced liver fibrosis (OR 11.057, 95% CI 1.193–102.439, p = 0.034) were independent risk factors of severe illness in COVID‐19 patients with NAFLD. Conclusion NAFLD patients with NFS‐determined advanced liver fibrosis are at higher risk of severe COVID‐19.


| INTRODUC TI ON
The coronavirus disease 2019 , caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has affected millions of individuals globally. The spectrum of COVID-19 ranges from asymptomatic disease to critical illness with fatal outcome. 1 Given the significant impacts of disease severity on prognosis of COVID, much interest has been paid to identify the risk factors associated with severe illness of COVID-19. Several factors, such as older age, pre-existing lung disease, diabetes mellitus, cardiovascular disease, and obesity, are associated with the severe illness of COVID-19. [2][3][4][5] Chronic liver disease (CLD) is another comorbid condition that is being evaluated for its impact on COVID-19 severity. It is reported that 2%-11% of patients with COVID-19 had comorbidities of CLDs. 6 Non-alcoholic fatty liver disease (NAFLD) is one of the most common CLDs worldwide. 7 The global prevalence of NALFD in general population is 6.3%-45%. 7,8 In most Asian countries including China, the prevalence of NAFLD is above 25%. 9 Our previous study demonstrated that 30.7% of COVID-19 patients had NAFLD. 10 We also found that COVID-19 patients with NAFLD are more likely to develop liver injury. 10 Ji et al. found that COVID-19 patients with fatty liver had a higher risk of disease progression, higher likelihood of abnormal liver function tests, and longer shedding time of SARS-CoV-2 virus. 11 Mushtaq et al also reported that NAFLD was an independent predictor of liver injury in hospitalized patients with COVID-19, while NAFLD was not identified as an independent predictor of mortality, disease severity, or disease progression. 12 NAFLD includes a wide spectrum of CLDs, from steatosis to non-alcoholic steatohepatitis, with a variable stage of fibrosis to cirrhosis. 13 The association of severity of liver fibrosis and the COVID- 19 is not yet clear. Targher et al. found that metabolic dysfunction-associated fatty liver disease (MAFLD) patients with increased FIB-4 are at higher risk of severe illness of COVID-19. 14 The NAFLD fibrosis score (NFS) was demonstrated to identify NAFLD patients with and without advanced fibrosis independently. 15 Nevertheless, at present, there is still little knowledge about the association of NFS with the severity of COVID-19 patients with NAFLD. We aimed to determine the association of NFS-determined liver fibrosis with clinical outcomes of COVID-19 in NAFLD patients.

| Study design and data collection
We retrospectively screened 342 consecutively confirmed COVID-19 patients who were admitted to hospitals between January 18, 2020, and February 26, 2020, from ten designated hospitals of Jiangsu Province, China. The patients were followed up to February 29, 2020. The COVID-19 was diagnosed according to the guidance provided by the World Health Organization (WHO). 16 Patients with viral hepatitis defined by positive serum hepatitis B surface antigen and/or hepatitis C antibody and/or a known history of chronic hepatitis B or chronic hepatitis C were excluded. Patients with a known history of autoimmune liver diseases, or any other chronic liver diseases were also excluded. In addition, we excluded patients with significant alcohol consumption defined by >30 g/day in men and >20 g/day in women. The epidemiological, clinical data and outcomes were collected from medical records of patients.
The study was approved by the Ethics Committees of these hospitals, with a waiver of informed consent.

| Definition of NAFLD, fibrosis, and severe illness
The NAFLD was diagnosed by the Hepatic Steatosis Index (HSI) in the absence of other causes of CLDs. 17 The HSI was calculated by the equation: HSI = 8 × (ALT/AST ratio) + BMI (+2, if female; +2, if diabetes). 17 Serum ALT and AST levels of the first test after patient admission were used for the calculation of the HSI. A cutoff of 36 of HSI was used to diagnose the presence of NAFLD. [18][19][20] The HSI has been validated and used in previous studies. [18][19][20] NAFLD fibrosis score was used to evaluate the severity of liver fibrosis. 15 NFS was calculated by the following formula: NFS = −1.675 + 0.037 × age (year) +0.094 × BMI (kg/m 2 ) +1.13 × IFG/diabetes (yes = 1, no = 0) + 0.99 × AST/ALT ratio − 0.013 × platelet count (×10 9 /L) -0.66 × albumin(g/dL). 15 Patients were separated into two groups: advanced liver fibrosis (ALF) and non-advanced liver fibrosis (non-ALF) according to NFS. 15 Conclusion: NAFLD patients with NFS-determined advanced liver fibrosis are at higher risk of severe COVID-19.

| Statistical analysis
Data were analyzed by SPSS version 22.0 software (SPSS Inc., Chicago, IL, United States). Continuous data were shown as medians and interquartile range (IQR), while categorical data were presented as the counts and percentages. Two-sample t tests or Mann-Whitney U test were used to compare the continuous variables between two groups as appropriate, while chi-square tests or Fisher's exact tests were used to compare the categorical variables. Logistic regression analysis was used to identify the risk factors of ALF. Variables having p values < 0.1 in the univariate analysis were used for a multivariate input logistic regression analysis. Age and gender were also adjusted by the multivariate logistic regression analysis. p < 0.05 was considered statistically significant.

| Laboratory and radiology examination
The baseline lymphocyte counts, platelet counts, and albumin levels were significantly lower in NAFLD patients with ALF compared with patients with non-ALF (all p < 0.001). All NAFLD patients with ALF had pneumonia in chest CT and most (92.1%) of the patients had bilateral pneumonia. However, 12.5% of the NALFD patients with non-ALF presented no pneumonia in chest CT. The proportion of pneumonia was higher in NAFLD patients with ALF than patients with non-ALF (p = 0.028) ( Table 2).

| Risk factors of severe illness during hospitalization
Logistic regression analysis was performed to identify the risk factors of severe illness in COVID- 19

| DISCUSS ION
The presence of NAFLD has been reported as a predictor of the development of liver injury. 11,12 In our previous study, we also demonstrated that COVID-19 patients with NAFLD are more likely to develop liver injury. 10  However, the sample size is very small which only included 38 patients. 23 Targher 15 A prospective study demonstrated that NFS had a more excellent discriminatory ability for the detection of advanced fibrosis than FIB-4. 26 Another study also reported that NFS had the highest discriminatory accuracy for advanced fibrosis compared with FIB-4. 27 However, there was no study investigating the association with NFS and the severity of COVID-19 in patients with NAFLD. In this study, we found that 44.2% of the NAFLD patients with COVID-19 had ALF according the NFS. Furthermore, we found that COVID-19 patients with NAFLD and increased NFS was an independent risk factor of severe illness of patients with COVID-19. We also found that the presence of diabetes was an independent risk factor of severe illness of COVID-19 which is consistent with the previous study. 28 The exact pathophysiology on how ALF contributes to a more severe COVID-19 illness in NAFLD patients is not yet clear, while several potential mechanisms have been proposed. Systemic inflammatory response syndrome, especially the "cytokine storm," is one of the important pathophysiologic mechanisms for the development of severe illness in COVID-19. 29 The presence of ALF might exacerbate the virus-induced "cytokine storm" and thereby contribute to severe COVID-19 through the hepatic release of proinflammatory cytokines. 14 However, more studies are needed to explore the mechanisms of advanced NAFLD contributing to the COVID-19 disease process.
Our study has some limitations. First, the sample size of our study is relatively small. Second, we use the non-invasive fibrosis scores to diagnose fibrosis without a histological diagnosis of liver fibrosis.
Other non-invasive tests such as transient elastography by FibroScan may be more reliable for the liver fibrosis and steatosis assessment. 30 However, during the early stages of the COVID-19 outbreak, the transient elastography was not performed in the patients. Third, we did not analyze the risk factors of ICU admission of COVID-19 patients with NAFLD since only 5 patients were admitted to the ICU.
Fourth, given that the quality of care may vary across hospitals, the results might be biased by the heterogeneity of care across hospitals.
However, all patients were treated according to the guidelines for the

Diagnosis and Treatment of coronavirus disease 2019 (COVID-19) is-
sued by the National Health Commission of China. 22 Thus, we consider the quality of care across hospitals might not significantly impact our results.
Our study demonstrates that patients with NAFLD with NFSdetermined ALF are at higher risk of having severe COVID-19.
However, more studies with large sample size are needed to validate

TA B L E 3 Risk factors of severe illness during hospitalization in COVID-19 patients
our results and to better understand the mechanistic link of advanced NAFLD and severe COVID-19.

CO N FLI C T S O F I NTE R E S T
The authors have declared that no conflicts of interest exist.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.