To the Editor:

Nonalcoholic fatty liver disease (NAFLD) is a burgeoning medical problem that affects 20%-34% of the population in Western countries.1 Although the prevalence of NAFLD is somewhat lower in Asia, the frequency is increasing, and the disorder is now being seen in younger individuals.2, 3

NAFLD is a multifactorial disorder. Major risk factors include obesity, insulin resistance, and a variation (rs738409) in patatin-like phospholipase domain-containing protein 3 (PNPLA3) that substitutes methionine for isoleucine at residue 148 (I148M).1 Only one prior study has examined the relationship between this variant and NAFLD in China. In that study, the variant was associated with fibrosis but not steatosis.4 Here, we examined the relationship between PNPLA3-I148M and liver triglyceride content in 203 unrelated adults with NAFLD who were recruited from an outpatient liver clinic at the First Hospital of China Medical University, Shenyang, China. Hepatic steatosis was diagnosed by liver ultrasonography using established criteria5; all other known causes of hepatic steatosis were excluded (see legend to Table 1). A total of 202 ethnically-matched controls with normal liver enzyme levels and no steatosis, as determined by ultrasonography, were recruited from primary care outpatient clinics at the same institution.

Table 1. Demographic and Clinical Characteristics of Subjects
CharacteristicNAFLD (n = 203)Control (n = 202)P Value
  1. Values are described by mean ± standard deviation for continuous variables; The median values ± interquartile range are provided for the BMI and triglyceride. NAFLD was diagnosed by ultrasonography.5 Secondary causes of steatosis were excluded (ethanol >30 g/day for men [>20 g/day for women], total parenteral nutrition, hepatitis B and hepatitis C virus, autoimmune liver disease, hemochromatosis, alpha1-antitrypsin deficiency, Wilson's disease, or use of drugs that promote steatosis). Nominal P values were calculated by a two-sample t test for continuous variables, a proportion test for categorical variables, and Fisher's exact test for genotypes ( PNPLA3 and APOC3 genotypes were assayed as described.1

  2. ALT, alanine aminotransferase; AST, aspartate aminotranferase; APOC3, apolipoprotein C3; BMI, body mass index; GGT, gamma-glutamyl transpeptidase; HDL-C, high density lipoprotein cholesterol; LDL-C, low density lipoprotein cholesterol; PNPLA3, patatin-like phospholipase domain-containing protein 3.

Age (years)46.6 ± 13.441.9 ± 13.11.80E-04
BMI (kg/m2) (median)26.6 ± 4.923.2 ± 4.32.10E-17
ALT (IU/L)44.3 ± 39.817.0 ± 8.01.00E-18
AST (IU/L)31.4 ± 28.019.8 ± 4.31.80E-08
GGT (IU/L)61.2 ± 16.124.8 ± 16.22.00E-04
Cholesterol (mmol/L)5.2 ± 1.04.5 ± 1.11.20E-09
Triglyceride (mmol/L)1.7 ± 1.21.0 ± 0.69.00E-18
LDL-C (mmol/L)3.3 ± 1.12.6 ± 1.14.30E-12
HDL-C (mmol/L)1.3 ± 0.61.9 ± 0.98.50E-14
Glucose (mmol/L)6.0 ± 1.35.6 ± 1.39.80E-04
PNPLA3-I148M allele0.450.311.50E-04
APOC3 C482T0.430.450.57
APOC3 T455C0.470.460.94
APOC3 combined variant0.650.660.92

After obtaining institutional review board approval and written informed consent from patients, fasting blood samples were collected. A higher proportion of cases were men (60% versus 49%; nominal P = 0.04) and their mean age was greater than that of controls (46.6 versus 41.9 years of age; P = 1.8 × 10−4). Levels of circulating liver enzymes, low density lipoprotein cholesterol (LDL-C), triglyceride, and glucose were significantly higher and high density lipoprotein cholesterol (HDL-C) levels were significantly lower in cases than in controls (Table 1).

The frequency of the I148M variant was significantly higher in cases (0.45) than in controls (0.31) (P = 1.5 × 10−4), and the association remained robust after adjusting for age, sex, and body mass index (P = 3.7 × 10−3) (Fig. 1). The odds ratio for hepatic steatosis was 1.73 for each copy of the G allele (95% confidence interval: 1.49, 1.99). The I148M variant was also associated with higher serum alanine aminotransferase levels (nominal P = 1.1 × 10−7, adjusted P = 2.0 × 10−6), but not with body mass index, fasting glucose, or with lipid/lipoprotein levels (data not shown). No association was found between NAFLD and two SNPs in the promoter region of APOC3 (Table 1).

thumbnail image

Figure 1. Genotype frequencies of rs738409 in cases with hepatic steatosis (n = 203) and controls (n = 202).

Download figure to PowerPoint

This is the first report to document an association between PNPLA3-I148M and hepatic steatosis in a population from mainland China. The relatively high frequency of the risk allele in China makes it imperative to control the weight gain that accompanies urbanization and adoption of Western eating habits.


  1. Top of page
  • 1
    Cohen JC, Horton JD, Hobbs HH. Human fatty liver disease: old questions and new insights. Science 2011; 332: 1519-1523.
  • 2
    Chitturi S, Wong VW, Farrell G. Nonalcoholic fatty liver in Asia: firmly entrenched and rapidly gaining ground. J Gastroenterol Hepatol 2011; 26( suppl 1): 163-172.
  • 3
    Fan JG, Peng YD. Metabolic syndrome and non-alcoholic fatty liver disease: Asian definitions and Asian studies. Hepatobiliary Pancreat Dis Int 2007; 6: 572-578.
  • 4
    Li X, Zhao Q, Wu K, Fan D. I148M variant of PNPLA3 confer increased risk for nonalcoholic fatty liver disease not only in European population, but also in Chinese population. HEPATOLOGY 2011; doi:10.1002/hep.24567.
  • 5
    Zeng MD, Fan JG, Lu LG, Li YM, Chen CW, Wang BY, et al. Guidelines for the diagnosis and treatment of nonalcoholic fatty liver diseases. J Dig Dis 2008; 9: 108-112.

Yiling Li M.D.* †, Chao Xing Ph.D.†, Jonathan C. Cohen Ph.D.†, Helen H. Hobbs M.D.† ‡, * Department of Gastroenterology, First Affiliated Hospital of China Medical University, Shenyang, China, † Eugene McDermott Center for Human Growth and Development, University of Texas Southwestern Medical Center, Dallas, TX, ‡ Howard Hughes Medical Institute, University of Texas Southwestern Medical Center, Dallas, TX.