We performed a familial aggregation study of fatty liver in overweight children with and without NAFLD. In family members, the presence or absence of fatty liver was documented using MRI. Clinical history and laboratory testing were applied to exclude participants with alternate explanations for fatty liver other than NAFLD. The main finding was that liver fat fraction and the condition of fatty liver are heritable traits. The estimate for the heritability of the fatty liver as a dichotomous trait reached the boundary value of 1.0, thus we were not able to determine the exact value for the heritability of fatty liver, but can confidently conclude that the heritability is high.
Earlier evidence for the familial nature of NAFLD came from retrospective case series. Struben et. al.
reviewed a clinical database of 124 patients with NASH and 174 with cryptogenic cirrhosis and identified 32 patients who reported a family history of fatty liver and/or cryptogenic cirrhosis. They reviewed the medical records of a total of 10 family members from 8 of the 32 patients with a positive family history and noted the co-existence of NASH and/or cryptogenic cirrhosis in seven of eight kindreds studied.6
Willner and colleagues reviewed the charts of 90 patients with a diagnosis of NASH at the University of Tennessee and the Medical University of South Carolina.7
The investigators noted that 16 of the 90 patients came from families with 2 or more patients with NASH. Notably cirrhosis was observed in 7 of these 9 families. A case series from Japan described 3 families each with 2 people with biopsy-proven NASH.26
In addition to familial aggregation of NAFLD, the phenotypic features associated with NAFLD may also be more prevalent in families with NAFLD. A study of 20 adults with NAFLD and 20 controls without NAFLD, demonstrated that insulin resistance was more prevalent in first degree relatives of patients with NAFLD than those without NAFLD.27
The frequency of fatty liver seen in the siblings and parents of overweight children without NAFLD was consistent with population-based estimates. The prevalence of fatty liver in children in the County of San Diego was estimated to be 9.6% overall and 17.3% for adolescents age 15 to 19. 1
For adults age 30 to 65, the prevalence of hepatic steatosis was estimated to be 34% in Dallas County.8
In sharp contrast, family members of children with NAFLD appear to be at high risk as the majority of both siblings and parents of children with NAFLD were shown to have hepatic steatosis. Targeted screening may be warranted. Moreover, a demonstration of the familial nature of the disease may serve to enhance efforts at modifying the family environment towards increased physical activity and healthier dietary choices. Of additional concern was the detection of 2 cases of unrecognized cirrhosis out of the 33 families with a child with NAFLD studied. Primary care physicians, endocrinologists, gastroenterologists, hepatologists, and pediatricians should consider the possibility of advanced liver disease in the absence of any symptoms.
A novel finding in the current study was the interaction between BMI and familial factors on liver fat fraction. In families of children with NAFLD (i.e., the families at higher risk for fatty liver), the severity of steatosis as determined by the MRI-measured fat fraction was strongly correlated with BMI. In contrast, in families of children without NAFLD (i.e., the families at lower risk for fatty liver), the correlation between BMI and liver fat fraction was weaker. This difference helps to explain why the adjustment for BMI attenuated the heritability estimate for liver fat fraction as a continuous trait but, by reducing noise, increased the heritability estimate for fatty liver as a dichotomous trait. Moreover, these observations suggest that BMI is a strong determinant of steatosis severity in those with a heritable susceptibility to fatty liver but not in those without a heritable susceptibility and helps to strengthen the argument that familial factors contribute to the development of hepatic steatosis.
Establishing the heritability of a trait is considered a first step towards the identification of specific genes that influence that trait. Although the strength of the heritability of a trait does not always translate into the ease with which genes can be identified, it does provide support that genes contribute to the phenotypic expression of that trait. It is likely that fatty liver that begins early in life has a stronger genetic component than fatty liver that does not develop until adulthood. Once heritability has been established, a logical next step is to pursue the identification of specific genes that influence the trait via linkage and association analysis.13
The accumulation of fat, largely composed of triglyceride, in the hepatocyte is the essential first step in the development of NAFLD. The pathogenesis of steatosis is likely multi-factorial, involving both complex genetic and environmental factors that regulate lipid metabolism and the flux of fatty acids to, within and from the hepatocyte.28, 29
Putative candidate genes include those involved in processes that regulate hepatic lipid metabolism and the flux of fatty acids to the liver from the adipocyte. Free fatty acid in the liver is metabolized by one of two pathways: oxidation to generate ATP or esterification to produce triacylglycerides. Triglycerides are incorporated into VLDL particles for export or stored within hepatocytes. Therefore, hepatic steatosis can be caused by any combination of factors that increase the hepatocyte free fatty acid pool or that decrease mitochondrial β-oxidation, peroxisomal γ-oxidation, or VLDL synthesis/secretion. Association studies of single nucleotide polymorphisms as a risk factor for NAFLD offer promising findings for genes related to: (1) the magnitude and pattern of fat deposition, (2) hepatic lipid export, and (3) inflammation and oxidant stress. 30–36
Similarly, studies of gene expression in hepatic steatosis have shown upregulation of genes involved in lipid metabolism.37–39
The studies to date are not definitive because of small sample sizes, the complex issue of appropriate control groups and the lack of replication samples. These data along with the current study support the planning and execution of large scale studies examining genetic factors contributing to NAFLD.
A strength of this study was the use of magnetic resonance imaging, as this allowed for objective and precise determination of hepatic fat fraction. Studies based upon less robust measures for case definition, such as liver chemistry or ultrasonography, are subject to greater error. The use of a diverse study population was notable, especially the inclusion of a large number of Hispanic families because they are at greater risk for NAFLD than non-Hispanics. However, this emphasis may limit the generalizability of these findings to non-Hispanic families. Another limitation was the lack of liver biopsies in the family members studied. A design including biopsy would be required to determine the hereditability of a sub-phenotype of NAFLD such as steatohepatitis or advanced fibrosis, but is not feasible for a family study which requires the use of a method that can be applied to all family members. Heritability estimates are influenced by shared environmental factors. In order to further separate environmental from genetic causes, future studies will need to collect information on factors including nutrition and activity.
The public health burden of NAFLD is considerable and likely to increase over time. Understanding the genetic and environmental factors that contribute to NAFLD prevalence and severity has important implications for clinical care and public health. The current data demonstrate for the first time that NAFLD is a highly heritable condition. Family members of children with NAFLD should be considered at high risk for NAFLD, even in the absence of obesity or elevated serum aminotransferases. The current data suggest that familial factors are a major determinant of whether an individual has NAFLD. Further family studies will help unravel the complex interactions between genes and environment in the development of NAFLD.