BACKGROUND: The influence of chronic hepatitis B infection (CBI) on hepatic steatosis, necroinflammation, and fibrosis in nonalcoholic fatty liver disease (NAFLD) population was unclear. We aimed to investigate the effect of CBI on hepatic steatosis and assess the association between NAFLD co-existed CBI and hepatic injury in NAFLD pediatric population.
METHODS: Consecutive hospitalized children with biopsy-proven NAFLD with or without CBI were included. Hepatic steatosis, necroinflammation and fibrosis were evaluated by NASH CRN system and/or METAVIR scoring system, appropriately. Using multivariate logistic analysis, we identified variables associated with hepatic steatosis and liver injury.
RESULTS: Of 223 biopsy-proven NAFLD children, 161 were NAFLD without CBI, and 62 were NAFLD co-existed CBI. Grouped by mild, moderate and severe hepatic steatosis, there was an inverse association between CBI and the severity of hepatic steatosis [odd ratio (OR) 0.037, 95% confidence interval (CI) 0.014-0.098]. In addition, we explored the relationship between CBI and hepatic necroinflammation and fibrosis in NAFLD children. Hepatic necroinflammation and fibrosis, respectively, were divided into two groups according to severity. And CBI was positively associated with hepatic necroinflammation (OR 6.125, 95%CI 1.958-19.158). However, there was no statistically independent association between CBI and significant hepatic fibrosis.
CONCLUSIONS: CBI was inverse associated with the grade of steatosis and positively associated with severe hepatic necroinflammation, and does not appear to affect significant hepatic fibrosis in pediatric NAFLD children.

BACKGROUND: The interaction between nonalcoholic fatty liver disease (NAFLD) and chronic hepatitis B infection (CBI) was unclear. We aimed to investigate the association between NAFLD and CBI and the effect of NAFLD on response to antiviral therapy in pediatric population.
METHODS: All children aged 0-18 years with liver biopsy-proven NAFLD, CBI, and co-existing NAFLD and CBI were consecutively collected. Children with co-existing CBI and NAFLD were considered as cases and n:m matched with simple NAFLD and simple CBI patients in the same cohort, respectively. In longitude study, the role of NAFLD in antiviral response was further analyzed in children with CBI who received antiviral treatment. Logistic or Cox regression models were used appropriately for analysis.
RESULTS: 765 subjects were finally enrolled with 62 co-existing patients, 560 CBI patients, and 143 NAFLD patients. Multivariate analysis showed that HBV DNA level was negatively associated with NAFLD in CBI children (OR 0.376, 95% CI 0.173-0.818). Conversely, the severity of steatosis and levels of serum lipid profile were found to be inversely associated with CBI in NAFLD subjects. Then, in longitude study, we found that HBsAg loss at 96 weeks of antiviral treatment was independently associated with NAFLD (aHR 3.245, 95% CI 1.288-8.176).
CONCLUSIONS: An inverse association between CBI and NAFLD reciprocally existed in pediatric population. In longitude study, HBsAg loss was associated with NAFLD at week 96 of antiviral therapy.