Hepatitis B and fatty liver: Causal or coincidental?

Over half a billion of the world's population is chronically infected with hepatitis B virus (HBV) and hepatitis C virus (HCV) infections. However, only a proportion of these individuals will develop progressive liver disease. It is therefore critical to identify which factors contribute to accelerated liver injury and a poor outcome. In patients with chronic hepatitis C (CHC), fat accumulation within the hepatocytes (hepatic steatosis, fatty liver) is one such variable that is associated with advanced hepatic fibrosis.1 Whether fatty liver in the context of chronic hepatitis B (CHB) has similar connotations has not been adequately clarified. In this editorial, we first examine the prevalence of hepatic steatosis, its probable origins and its potential clinical implications in HBV-infected individuals. Fatty liver refers to non-alcoholic fatty liver disease (NAFLD); alcohol-related steatosis is not discussed here. Of the studies published to date (Table 1), hepatic steatosis is observed in at least one-quarter (38%, 338 of 899 cases) of individuals with CHB.2–9 In studies which included subjects with CHB and also CHC with comparable metabolic profiles, the overall prevalence of fatty liver is lower in CHB-infected individuals than in persons with CHC (approximately 50%).1 A few caveats apply. In some studies, hepatic steatosis was defined by hepatic ultrasound and not by liver histology. In others, persons with significant alcohol intake were not excluded. Not withstanding these caveats, based on current global trends with obesity and type 2 diabetes, it seems certain that the number of patients with CHB and fatty liver is set to rise. As we discuss next, host characteristics such as obesity and type 2 diabetes are important determinants of hepatic steatosis in CHB. In persons with genotype 3 CHC, there is general consensus that hepatic steatosis is directly attributable to the hepatitis C virus, possibly by enhanced fatty acid uptake from lipoproteins. In genotypes 1 and 4 HCV infections, insulin resistance and oxidative stress may also promote de novo fatty acid biosynthesis, impair mitochondrial β-oxidation, and disrupt triglyceride assembly and secretion.1,10 Is there a similar role for HBV in patients with CHB and fatty liver? There are experimental data to support this hypothesis. Recently, Kim et al. showed that increased HBV X protein expression can induce lipid accumulation in hepatocytes. Hepatic lipogenesis was mediated by upregulation of sterol regulatory element binding protein 1 (SREBP-1) and peroxisome proliferator activated receptor gamma (PPARγ).11 However, clinical data do not support this concept of HBV-induced steatosis. In particular, hepatic steatosis does not correlate with hepatitis B e antigen (HBeAg) status or HBV-DNA titer.2–9 Genotype-specific studies have yet to be conducted, but the concept of ‘steatogenic’ HBV genotypes (as for HCV) remains speculative. As viral characteristics do not appear to be critical, it is therefore important to consider whether the hepatic steatosis is attributable to host factors. After all, in patients with genotypes 1 and 4 CHC, the prevalence of fatty liver appears to be dictated by the host metabolic profile, as shown by its strong association with insulin resistance, type 2 diabetes, body mass index (BMI) and components of the metabolic syndrome.1,10 In this issue of the Journal, Wang et al. from Taiwan examined the determinants of hepatic steatosis and insulin resistance in persons with chronic HBV infection.9 The authors reviewed the health examination records of 507 adults for markers of hepatitis B surface antigen (HBsAg). All persons were negative for antibody against HCV and consumed less than 140 g alcohol/week. Among the 507 adults, 50 subjects (9.9%) had evidence of HBV infection, and 28 of these (56%) had findings of ultrasonographic fatty liver. Their anthropometric and biochemical characteristics were compared with subjects without HBV infection (n = 457). The results showed clearly that the indices of insulin resistance (HOMA-IR) were similar in both groups. Further, in multivariate logistic regression analyses, positive HBsAg status was also not identified as an independent predictor of fatty liver or of insulin resistance. However, the investigators found strong correlations between hepatic steatosis and insulin resistance, BMI and triglyceride levels, as would be expected of HBsAg-negative subjects with NAFLD. These observations are in line with other reports in patients with CHB that show an association between steatosis and insulin resistance, with its clinical concomitants of obesity/central obesity, hyperglycemia, hypertriglyceridemia, hypertension, and low high-density-lipoprotein–cholesterol levels (Table 1).2–9 It is therefore reasonable to conclude that host factors underlie intrahepatic fat accumulation in these individuals. Furthermore, unlike CHC, chronic HBV infection is not a disorder with heightened insulin resistance. To the contrary, chronic HBV infection may even be protective! Thus, two recent studies from China and Taiwan, respectively, have shown inverse relationships between the prevalence of metabolic syndrome and positive HBsAg status.12,13 These results need to be replicated and, if found to be reproducible, the mechanistic explanation may be of considerable interest. Only limited, retrospective data are available addressing these issues. Thus far, the grade of hepatic steatosis has not been shown to correlate either with hepatic necroinflammatory activity or fibrosis stage in CHB. Longitudinal studies that have examined the impact of fatty liver on the natural history of CHB are lacking. However, the lack of association with fibrotic severity, a predictor of poor outcome in CHB, suggests that the coexistence of hepatic steatosis is, by itself, unlikely to have negative consequences. Conversely, NAFLD is not altogether a benign disorder. Up to 5% of those affected may develop cirrhosis, and liver-related deaths can occur in this subgroup. Further, diabetes, a key risk factor for NAFLD, is associated with increased risk of hepatocellular carcinoma, irrespective of underlying disease etiology, including CHB.14 In the future, the individual impact of these disorders (steatosis, CHB) could still be significant. To date, however, all that can be concluded is that there is no obvious synergism between these two entities as in the case of CHC. Steatosis does not appear to negatively impact the treatment outcome of individuals with CHB. The outcome of treating patients using a pegylated interferon-based regimen did not significantly differ between groups of CHB patients with and without hepatic steatosis.7 This issue has not been looked at specifically with reference to the nucleos(t)ide analogs, but the presence of steatosis was not flagged as a negative predictor of successful antiviral treatment in the registration trials with these drugs. Finally, a seemingly unrelated but pertinent observation is that, in Taiwan, CHB-infected persons with moderate-to-severe steatosis were more than threefold likely to achieve spontaneous HBsAg seroclearance than those without steatosis.15 In the context of the hotly debated obesity-diabetes-hepatocellular carcinoma link, this report muddies the water further! Urgent reconfirmation is needed because of its obvious implications for individuals with CHB. In conclusion, hepatic steatosis in persons with CHB is a reflection of the host metabolic profile. At present, there is no evidence that it has a negative impact on the clinical course of CHB, and it does not influence the outcome of antiviral treatment. CHB is not in itself correlated with insulin resistance. However, in persons with inactive or low-level HBV infection who also have fatty liver and coexisting insulin resistance-related disorders, it is still prudent to seek strategies aimed at reducing the liver fat content to prevent potentially hazardous hepatic, metabolic and cardiovascular complications.16