The association between serum vitamin A concentrations and virus hepatitis among US adults from the NHANES database: a cross-sectional study

Objective: According to the present study, the relationship between vitamin A (VA) levels and hepatitis virus carriage has been unclear and controversial. This study aimed to determine the potential relationship between serum VA levels and viral hepatitis and to provide ideas for future clinical treatments. Methods: A cross-sectional study was performed using the 2005-2006 and 2017-2018 National Health and Nutrition Examination Survey (NHANES) datasets. Multiple linear regression and logistic regression were adopted to analyze the association between serological hepatitis B surface antigen (HBsAg) or hepatitis C RNA (HCV-RNA) positivity and VA levels. There were 5,351 HBsAg-related responders and 242 HCV-RNA-related responders, including 52 HBsAg (+) and 104 HCV-RNA (+) responders. Results: Compared with HBsAg (-) and HCV-RNA (-) respondents, HBsAg (+) and HCV-RNA (+) respondents tended to have lower serum VA levels, respectively [1.63 (1.33 similar to 2.01) vs. 1.92 (1.57 similar to 2.34), P < 0.001; 1.54 (1.25 similar to 1.83) vs. 1.78 (1.46 similar to 2.26), P < 0.001]. A greater percentage of responders in the subclinical VA deficiency (SVAD) group were HBsAg (+) and HCV-RNA (+) than were those in the normal VA (VAN) group [2.4% (9/374) vs. 0.9% (43/4977), p = 0.003; 61.5% (16/26) vs. 40.7% (88/215), p = 0.043]. According to the results of the multiple regression analyses of the different models, the serum VA concentration was negatively correlated with HBsAg (+) and HCV-RNA (+) status (beta = -0.14, 95% CI = -0.30 to -0.01, p = 0.066; beta = -0.29, 95% CI = -0.50 similar to -0.09, p = 0.005, respectively). Compared to those with SVAD, patients with VAN were less likely to be serologically HBsAg (+) or HCV-RNA (+) (OR = 0.53, 95% CI = 0.25 similar to 1.10, p = 0.089; OR = 0.39, 95% CI = 0.18 similar to 0.84, p = 0.016, respectively). Conclusion: Our study provides evidence that patients who are HBsAg (+) or HCV-RNA (+) have a high incidence of SVAD. Moreover, HBsAg and HCV-RNA positivity are negatively correlated with VA levels, and patients with SVAD are more likely to carry HBsAg (+) or HCV-RNA (+). These findings suggest that the relationship between hepatitis viruses and vitamin A needs to be validated by more basic studies and clinical large-sample randomized controlled trials to provide ideas for new therapeutic targets.