William F. Balistreri, MD


April 05, 2016

In This Article

Editor's Note:
Fatty liver disease is one of the major liver-related challenges that faces clinicians. Significant advances in the diagnosis and management of patients with this increasingly common and daunting disease were presented at The Liver Meeting® 2015—the 66th annual meeting of the American Association for the Study of Liver Diseases (AASLD). Some of the presentations from this meeting are highlighted herein to assist clinicians in ensuring optimal outcomes for their patients.

Nonobese Fatty Liver Disease

Recognition and management of patients with nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH) is a vexing clinical challenge.

Although the characteristic "phenotype" of patients with NAFLD is one of central adiposity, a proportion of patients with NAFLD have normal body mass index (BMI)—something that we might refer to as "nonobese fatty liver disease." Studies suggest that such patients may have more severe disease and worse prognoses, though these findings are potentially limited by selection bias.

Wong and colleagues[1] studied the epidemiology and severity of nonobese NAFLD in community-based persons randomly recruited from the census database of the Hong Kong government. An intrahepatic triglyceride content ≥ 5.0% by proton-magnetic resonance spectroscopy was the threshold used to define NAFLD. Serum cytokeratin-18 fragments were measured to reflect hepatocyte apoptosis and steatohepatitis. Liver stiffness was measured by transient elastography (FibroScan), and the NAFLD fibrosis score was used to assess liver fibrosis.

Among 911 recruited persons, 77% had a BMI < 25 kg/m2 and 23% had a BMI ≥ 25 kg/m2. The prevalence of NAFLD was 19% in nonobese persons and 61% in obese persons (P < .001). Among those with NAFLD, nonobese patients had similar intrahepatic triglyceride levels to their obese counterparts, but lower cytokeratin-18 fragments, liver stiffness, and NAFLD fibrosis score. The G allele at the patatin-like phospholipase domain-containing protein 3 gene (PNPLA3 rs738409) was more common in nonobese patients with NAFLD (78% vs 60%). Increased waist circumference, insulin resistance, reduced high-density lipoprotein (HDL) cholesterol levels, elevated plasma triglyceride and ferritin levels, and the PNPLA3 G allele were independent factors associated with NAFLD in non-obese subjects. Even among non-obese persons with normoglycemia, those with NAFLD were more likely to be insulin-resistant. Only four nonobese patients had increased liver stiffness suggestive of advanced fibrosis; three of them had the metabolic syndrome, and all four carried the PNPLA3 G allele.

The observation that nearly 20% of the general nonobese Chinese population has NAFLD has important implications. Nonobese patients with risk factors for advanced fibrosis, such as the metabolic syndrome and presence of the PNPLA3 G allele, should be further assessed.

Distinct Phenotypes of NAFLD

The distribution of steatosis in patients with NAFLD can be categorized as panacinar (even distribution of fat droplets throughout), zone 1 (periportal), zone 3 (perivenular), or azonal (unable to determine, or focal sparing with zonal location).

Focal zone 1 pathology, although rare in adults with NAFLD (< 1%), has been reported in children. Africa and coworkers[2] hypothesized that focal zone 1 steatosis and focal zone 3 steatosis are two distinct subphenotypes of childhood NAFLD. They therefore determined the association between the zonality of steatosis and clinical, demographic, and histologic features in children enrolled in the NASH Clinical Research Network (NASH CRN).

The most common distribution of steatosis was panacinar (40%), with zone 1 steatosis present in 18% and zone 3 steatosis in 31%. Children with zone 1 steatosis were younger than those with zone 3 steatosis. There was no difference in steatosis distribution by sex. Compared with children who had zone 3 steatosis, children with zone 1 steatosis were significantly more likely to be Hispanic (73% vs 58%) and to have portal inflammation (97% vs 84%), fibrosis (81% vs 50%), and advanced fibrosis (13% vs 5%). In contrast, children with zone 3 steatosis were significantly more likely to have ballooning (48% vs 29%) and steatohepatitis (30% vs 6%).

These data suggest that zone 1 and zone 3 distribution of steatosis represent important subphenotypes of NAFLD in children. The authors discuss the paradox that at diagnosis, children with zone 1 steatosis are more likely to have advanced fibrosis but less likely to have steatohepatitis. Thus, the long-term risk for cirrhosis may also differ between children with zone 1 and those with zone 3 steatosis. To achieve a comprehensive understanding of NAFLD, studies of pathophysiology, natural history, and response to treatment should account for the zonality of steatosis.

Proposing a New Balloon Score

A criticism of the NAFLD Activity Score (NAS) has been that it gives less weight to ballooning than to steatosis or lobular inflammation.

Kleiner and colleagues[3] prospectively classified all biopsies with ballooning as classic (distinct, enlarged, with clumped cytoplasm) or nonclassic (cytoplasmic changes without distinctive cytomegaly). A new NASH CRN balloon score was created by incorporating these characteristics into the old balloon score.[3]

The investigators reviewed the NASH CRN database (1249 patients) and observed that—compared with cases without ballooning—nonclassic cases had more fibrosis, steatosis, portal and lobular inflammation, and were more likely to have the metabolic syndrome and elevated fasting insulin levels (all P < .01). Cases with severe ballooning had more fibrosis, portal and lobular inflammation, and higher alanine aminotransferase (ALT) levels (all P < .002). Age, female sex, BMI, diabetes, metabolic syndrome, ALT, fasting insulin and glucose levels, and homeostatic model assessment of insulin resistance (HOMA-IR) were all positively associated with the new balloon score.

The authors suggest that this revised balloon score, which shows excellent correlation with clinical disease features, should replace the old balloon score in the NAS. This would give more weight to ballooning, as well as increase the dynamic range of the NAS.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.