Hepatology Research 2017; 47: E104–E112 doi: 10.1111/hepr.

12734

Original Article

Relationship of serum uric acid level with non-alcoholic fatty

liver disease and its inflammation progression in non-obese
adults
Jing Liu,1 Chengfu Xu,2 Limei Ying,3 Shufei Zang,4 Zhenjie Zhuang,5 Haifeng Lv,6 Wenjun Yang,7
Yan Luo,5 Xaojie Ma,1 Lei Wang,8 Yunhao Xun,9 Dewei Ye10 and Junping Shi1†
Departments of 1Liver Diseases, 4Endocrine Diseases, 5Center for Translational Medicine, and 7Pathology, The
Affiliated Hospital of Hangzhou Normal University, 2Department of Gastroenterology, The First Affiliated Hospital,
College of Medicine, Zhejiang University, 8Second Clinical Medical College, Zhejiang Chinese Medicine University,
and 9Department of Liver Diseases, Xixi Hospital of Hangzhou, Hangzhou, Zhejiang, 3Dalian University of Medicine,
Dalian, Liaoning and 10Department of Medicine, Faculty of Medicine, The University of Hong Kong, and 6Department
of Surgical Intensive Care Unit, The First Affiliated Hospital of Zhejiang University, China

Aim: This study aimed to evaluate the relationship between se-
rum uric acid (SUA) level and non-alcoholic fatty liver disease
(NAFLD) in non-obese adults.
Methods: A cross-sectional study was carried out among 4098
adults, including 1936 non-obese and 2162 obese individuals. An
additional 93 non-obese adults with biopsy-proven NAFLD were
also included.
Results: The overall prevalence of NAFLD was 39.51% in the
study group, and 14.88% in non-obese adults. The NAFLD pa-
tients had significantly higher SUA levels than controls in both
men and women. The non-obese group had a higher NAFLD risk
with increased SUA levels than the obese group, with odd ratios
(95% confidence interval) of 2.559 (1.870–3.503) and 1.692
(1.371–2.087), respectively. In 93 non-obese adults with biopsy-
proven NAFLD, SUA levels were significantly higher in those with
non-alcoholic steatohepatitis. The prevalence of non-alcoholic
steatohepatitis and lobule inflammation tended to increase to
57.58% and 66.67% as the SUA level increased to the fourth quar-
tile. Subjects with hyperuricemia had significantly higher NAFLD
activity scores and more serious lobule inflammation than the
normal group.
Conclusion: Non-obese adults have higher NAFLD risk with in-
creased SUA levels than obese individuals, and the inflammation
progression of NAFLD is associated with increased SUA level in
non-obese subjects.
Key words: inflammation progression, non-obese adults, non-
alcoholic fatty liver disease, uric acid

INTRODUCTION Europe and North America, but also becoming increas-

ingly prevalent in the Asia-Pacific region, especially in
N ON-ALCOHOLIC FATTY LIVER disease (NAFLD) is
characterized by significant lipid deposition, usually
greater than 5% of the liver weight deposited as triglycer-
China with the prevalence of 15–30% in the general pop-
ulation.2 Non-alcoholic fatty liver disease mainly occurs in
obese individuals, but recent studies revealed that NAFLD
ide, in the liver parenchyma, without history of excessive
is also not rare in non-obese adults.3 The prevalence of
alcohol consumption. The NAFLD spectrum ranges from
NAFLD in non-obese adults was 15.2% in a Japanese study
simple steatosis (non-alcoholic fatty liver), to non-
group.4 So it is time to pay attention to NAFLD in non-
alcoholic steatohepatitis (NASH), to cirrhosis, ultimately
obese adults.
developing into hepatocellular carcinoma.1 Nowadays,
There is growing evidence that serum uric acid (SUA)
NAFLD is not only the most common liver disease in
level, the final product of purine degradation, is elevated
in metabolic syndrome (MetS) and strongly associated
Correspondence: Professor Jun-ping Shi, Department of Liver Diseases, The
with insulin resistance. The level of SUA is maintained
Affiliated Hospital of Hangzhou Normal University, 126 Wenzhou Road,
Hangzhou 310015, Zhejiang, China. Email: 13957121199@126.com by the balance between uric acid production and excre-
Conflict of interest: The authors have no conflict of interest. tion in physiological conditions. However, the preva-
Received 6 February 2016; revision 12 April 2016; accepted 26 April 2016. lence and incidence of hyperuricemia in the world

© 2016 The Japan Society of Hepatology E104

Hepatology Research 2017; 47: E104–E112
population have steadily increased over the past
40 years.5 Although uric acid is recognized as an antiox-
idant, hyperuricemia has been implicated in the develop-
ment of gout, coronary artery disease, type 2 diabetes,
MetS, and NAFLD.6–9 Increased SUA has been linked to
increased oxidative stress, chronic low grade inflamma-
tion, and insulin resistance,10,11 which are the basic
pathophysiology of NAFLD.
Although numerous studies have reported that SUA
level is associated with NAFLD,12,13 no study has aimed
to investigate the association of SUA level with NAFLD
and its inflammation progression in non-obese adults.
This study aimed to investigate the clinical and histopath-
ologic characteristics of NAFLD, and then analyze the rela-
tionships between SUA, NAFLD, and degree of
inflammation in non-obese adults.
MATERIALS AND METHODS
Study design and subjects
A CROSS-SECTIONAL STUDY was carried out to evalu-
ate the relationship between SUA level and NAFLD.
This study initially enrolled all of the adults who took their
health examination at Xixi Hospital (Hangzhou, China)
and the Affiliated Hospital of Hangzhou Normal Univer-
sity (Hangzhou, China) between January 1, 2013 and De-
cember 31, 2014. Subjects meeting the following criteria
were excluded: (i) those taking antihypertensive or antidi-
abetic agents, lipid-lowering agents, or hypouricemic
agents; (ii) those with alcohol consumption greater than
140 g/week for men and 70 g/week for women; (iii) those
with a history of other known causes of chronic liver dis-
ease, such as viral hepatitis or autoimmune hepatitis, and
(iv) those using hepatotoxic medications. A total of 4098
eligible subjects were enrolled. An additional 93 non-
obese adults with biopsy-proven NAFLD were also in-
cluded. The study protocol was approved by the ethics
committee, complying with the ethical guidelines of the
Declaration of Helsinki in 1995 (as revised in Edinburgh
2000). And written informed consent was obtained from
all subjects.
Clinical examination and biochemical analyses
All participants underwent routine medical history and
physical examination including blood pressure measure-
ment, anthropometry, and laboratory assessments. Body
mass index (BMI) was calculated as weight (kg) divided
by height (m) squared (kg/m2) and used as an index of
body fat. Clinical examinations were undertaken in the
morning after an overnight fast, and subjects were also
instructed to refrain from exercise during the day prior
Serum uric acid level E105
to their examination. The examination consisted of a
physical examination and a health habit inventory by a
physician.
Fasting blood samples were obtained from an
antecubital vein, and the samples were used for the
analysis of biochemical values. The values included al-
anine aminotransferase (ALT), aspartate aminotransfer-
ase (AST), triglyceride (TG), total cholesterol (TC),
high-density lipoprotein cholesterol (HDL-C), low-
density lipoprotein cholesterol (LDL-C), c-
glutamyltransferase, SUA, and fasting plasma glucose
(FPG). All values were measured by an automatic bio-
chemical analyzer (model 7180; Hitachi, Tokyo, Japan)
using standard methods.
Ultrasonography
Hepatic ultrasonic examination was carried out in all
subjects by two trained ultrasound doctors who were
blinded to the clinical and laboratory data, using a
Toshiba Nemio 20 sonography machine (Toshiba, To-
kyo, Japan) with a 3.5-MHz probe. Hepatic steatosis
was diagnosed by characteristic echo patterns according
to conventional criteria, such as the evidence of diffuse
hyperechogenicity of the liver relative to the kidneys, ul-
trasound beam attenuation, and poor visualization of
intrahepatic structures.14
Liver histology
The histopathological diagnosis of NAFLD was
established by two expert liver pathologist, blinded to
subjects’ details. Liver biopsy specimens were scored
using the NASH CRN scoring system.15 In brief, the de-
gree of steatosis, liver injury, and inflammatory activity
were scored using an 8-point scale: steatosis, 0–3; lobular
inflammation, 0–3; ballooning hepatocyte degeneration,
0–2. The stage of fibrosis was scored using a 6-point
scale: 1a, mild zone 3 perisinusoidal fibrosis; 1b, moder-
ate zone 3 perisinusoidal fibrosis; 1c, portal fibrosis only;
2, zone 3 and portal/periportal fibrosis; 3, bridging fibro-
sis; 4, cirrhosis. In addition to determining the NAFLD
activity score (NAS), an overall diagnostic categorization
was determined for each case as simple steatosis (NAS,
0–2), indefinite NASH (NAS, 3–4), and NASH (NAS,
5–8). According to the fibrosis stage, cases were divided
into non-advanced fibrosis (stage of fibrosis, 0–2) and
advanced fibrosis.
Definitions and statistical analyses
Non-alcoholic fatty liver disease was diagnosed by abdom-
inal ultrasound following exclusion of alcohol consump-
tion and viral or autoimmune liver disease.16,17 A BMI of
© 2016 The Japan Society of Hepatology
E106 J. Liu et al. Hepatology Research 2017; 47: E104–E112

≥18.5 kg/m2 and <25 kg/m2 for non-obese subjects, and
BMI ≥25 kg/m2 for obese subjects was defined according
to the World Health Organization guidelines of categoriz-
ing obesity in Asian populations as proposed in 2000. Hy-
peruricemia was defined as an SUA level >420 μmol/L in
male subjects and >360 μmol/L in female subjects. For
men, quartiles (Q) were defined as: Q1,
SUA ≤ 309 μmol/L; Q2, SUA 310–353 μmol/L; Q3, SUA
354–401 μmol/L; and Q4, SUA ≥ 402 μmol/L. For women,
quartiles were defined as: Q1, SUA ≤ 217 μmol/L; Q2, SUA
218–253 μmol/L; Q3, SUA 254–298 μmol/L; and Q4,
SUA ≤ 299 μmol/L.
Statistical analyses were carried out using the SPSS
software package version 13.0 for Windows (SPSS Inc.,
Chicago, IL, USA). Continuous variables are presented as
the mean and standard deviation (SD) or the median
and interquartile range (IQR), as appropriate. Student’s
t-test or the Mann–Whitney U-test was used for two-group
comparisons of continuous data; categorical variables were
compared using the χ 2-test. One-way ANOVA followed by
Dunnett’s test was used for multiple group comparisons.
Linear correlation analysis was used to determine the rela-
tionship between SUA level and other parameters of
NAFLD patients. Multivariate logistic regression (back-
ward Wald; entry, 0.05; removal, 0.10) was used to evalu-
ate the risk factors for NAFLD. P < 0.05 was considered
statistically significant.
RESULTS
Clinical and laboratory characteristics of
subjects
O F 4098 SUBJECTS enrolled in this study, 1619
(39.51%) fulfilled NAFLD diagnostic criteria. An-
thropometric, clinical, and laboratory data of 4098 sub-
jects are shown in Table 1. Compared with healthy
subjects, NAFLD patients were older, had significantly
higher BMI, FPG, TG, TC, LDL-C, ALT, AST, serum creati-
nine (SCr), blood urea nitrogen (BUN), and SUA levels,
and lower HDL-C levels.
There were 1936 non-obese subjects
(18.5 ≤ BMI < 25 kg/m2) and 2162 obesity subjects
(BMI ≥ 25 kg/m2). The prevalence of NAFLD was 14.88%
(288/1936) in the non-obese group, and 61.56% in the
obese group. In the non-obese study group, NAFLD pa-
tients were older, and they had significantly higher BMI,
FPG, TG, TC, LDL-C, ALT, AST, SCr, and BUN levels, and
lower HDL-C levels than controls. Notably, significantly
higher SUA levels were observed in subjects with NAFLD
than in those without NAFLD in the non-obese group
(men, 368.00 ± 75.45 vs. 332.72 ± 72.82 μmol/L; women,
272.00 ± 71.84 vs. 232.46 ± 54.72 μmol/L) (Fig. 1a). Com-
pared with NAFLD patients in the obese group, non-obese
patients were older and had significantly lower BMI, TG,
ALT, SCr, and SUA levels (Table 2). In female subjects,

Table 1 Clinical and laboratory characteristics of 4098 adults, including 1619 adults with non-alcoholic fatty liver disease (NAFLD)

Variable Total Healthy group NAFLD group P-value

n (women) 4098 (1515) 2479 (1096) 1619 (419) 0.000†

Age, years 41.73 ± 11.62 39.38 ± 11.34 45.52 ± 11.07 0.000
BMI, kg/m2 25.31 ± 3.94 23.62 ± 3.23 28.02 ± 3.45 0.000
SUA, μmol/L 316.63 ± 97.11 294.15 ± 93.94 353.08 ± 90.95 0.000
FPG, mmol/L 5.31 ± 1.32 5.12 ± 1.08 5.62 ± 1.60 0.000
TG, mmol/L 1.34 (0.88–2.08) 1.07 (0.75–163) 1.89 (1.31–2.82) 0.000‡
TC, mmol/L 4.69 ± 0.89 4.50 ± 0.92 5.00 ± 1.01 0.000
HDL-C, mmol/L 1.24 ± 0.37 1.31 ± 0.37 1.13 ± 0.35 0.000
LDL-C, mmol/L 2.88 ± 0.79 2.74 ± 0.75 3.11 ± 0.80 0.000
ALT, U/L 20.70 (14.90–30.40) 17.60 (13.00–24.90) 27.00 (19.00–39.00) 0.000‡
AST, U/L 19.00 (16.00–23.40) 18.00 (15.50–22.00) 21.00 (17.70–26.00) 0.000‡
BUN, mmol/L 5.09 ± 1.48 4.97 ± 1.43 5.29 ± 1.54 0.000
SCr, mmol/L 71.36 ± 19.58 69.80 ± 16.91 73.89 ± 23.04 0.000

Data are expressed as mean ± standard deviation or median (interquartile range).

‡P-value calculated using the Mann–Whitney U-test.
2
†P-value calculated using the χ -test.
ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; BUN, blood urea nitrogen; FPG, fasting plasma glucose;
HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; SCr, serum creatinine; SUA, serum uric acid; TC, total
cholesterol; TG, triglyceride.

© 2016 The Japan Society of Hepatology

Hepatology Research 2017; 47: E104–E112 Serum uric acid level E107

Figure 1 Serum uric acid (SUA) level distribution according to sex and age of study participants. (a) SUA levels in men and women.
*P < 0.05 versus non-obese non-alcoholic fatty liver disease (NAFLD) absent; ‡P < 0.05 versus non-obese NAFLD present. (b) SUA level
separated by age in female adults. *P < 0.05 versus their corresponding group aged <45 years; ‡P < 0.05 versus their corresponding
group aged >55 years. , Nonobese NAFLD Absent; , Nonobese NAFLD Present; , Obesity NAFLD Present; , Obesity NAFLD
Absent; , Nonobese NAFLD Absent; , Nonobese NAFLD Present; , Obesity NAFLD Present; , Obesity NAFLD Absent.

Table 2 Clinical and laboratory characteristics of adults with non-alcoholic fatty liver disease (NAFLD) according to body mass index (BMI)

Variable Non-obese subjects (n = 1936 ) Obese subjects (n = 2162 )

NAFLD absent NAFLD present P-value NAFLD absent NAFLD present P-value

n (women) 1648 ( 871 ) 288 ( 99 ) 0.001† 831 ( 225 ) 1331( 320 ) 0.026†
Age, years 37.89 ± 11.22 46.86 ± 10.99 0.000 42.50 ± 10.96 45.16 ± 11.03‡ 0.000
BMI, kg/m2 21.87 ± 1.99 23.56 ± 1.27 0.000 27.26 ± 2.04 28.93 ± 2.93‡ 0.000
SUA, μmol/L 279.13 ± 81.05 335.15 ± 87.03 0.000 326.12 ± 90.90 356.78 ± 90.90‡ 0.000
FPG, mmol/L 5.06 ± 1.07 5.53 ± 1.68 0.000 5.25 ± 1.09 5.64 ± 1.57 0.000
TG, mmol/L 0.96 (0.69–1.41) 1.77 (1.30–2.58) 0.001§ 1.41 (0.97–2.06) 1.93 (1.31–2.87)‡ 0.000§
TC, mmol/L 4.38 ± 0.90 5.01 ± 1.68 0.000 4.77 ± 0.94 5.00 ± 0.99 0.000
HDL-C, mmol/L 1.37 ± 0.37 1.16 ± 0.31 0.000 1.19 ± 0.33 1.13 ± 0.35 0.001
LDL-C, mmol/L 2.60 ± 0.72 3.01 ± 0.80 0.000 3.02 ± 0.74 3.11 ± 0.80 0.010
ALT, U/L 16.00 (12.00–21.85) 20.00 (17.00–23.40) 0.001§ 19.40 (16.20–23.62) 28.07 (20.00–40.00)‡ 0.000§
AST, U/L 17.80 (15.00–21.00) 23.65 (17.95–31.93) 0.001§ 22.35 (16.76–31.06) 21.20 (17.80–26.90)‡ 0.000§
BUN, mmol/L 4.88 ± 1.46 5.32 ± 1.48 0.000 5.14 ± 1.35 5.28 ± 1.55 0.033
SCr, mmol/L 67.59 ± 16.42 70.68 ± 16.37 0.009 75.03 ± 16.78 74.61 ± 26.36‡ 0.630

Data are expressed as mean ± standard deviation or median (interquartile range). Calculated by comparison of subjects with NAFLD within each
BMI group. Significant values appear in boldface type.
2
†P-value calculated using the χ -test.
‡For comparison of non-obese subjects and obese subjects with NALFD, P < 0.05.
§P-value calculated using the Mann–Whitney U-test.
ALT, alanine aminotransferase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; FPG, fasting plasma glucose; HDL-C, high-density li-
poprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; SCr, serum creatinine; SUA, serum uric acid; TC, total cholesterol; TG, triglyceride.

SUA level increased with age in both non-obese and obese
groups. Among non-obese NAFLD patients, women after
menopause had significantly higher SUA levels than
women before menopause (286.44 ± 63.60 vs. 254.96
± 70.88 μmol/L) (Fig. 1b).
Association between SUA and other clinical
parameters
Linear correlation analysis was used to determine the re-
lationship between SUA level and other clinical
parameters. Eleven variables including age, BMI, TG,
TC, HDL-C, LDL-C, ALT, AST, FPG, SCr, and BUN were
involved (Table 3). There were extremely significant pos-
itive correlations between BMI, TG, TC, LDL-C, ALT, AST,
SCr, BUN and SUA, whereas HDL-C had negative corre-
lations in both non-obese and obese subject groups. Re-
markably, compared with the obese group, the
correlation coefficients between BMI, TG, TC, LDL-C,
HDL-C, ALT, SCr, BUN, and SUA were all higher in the
non-obese group.

© 2016 The Japan Society of Hepatology

E108 J. Liu et al. Hepatology Research 2017; 47: E104–E112

Table 3 Association between serum uric acid (SUA) level and NAFLD in non-obese subjects. Remarkably, the non-obese
other parameters in non-obese and obese adults group had higher NAFLD risk with increased SUA levels
Variable Non-obese subjects Obese subjects than the obese group (odds ratio, 2.559 and 95% confi-
(n = 2038 ) (n = 2164 ) dence interval, 1.870–3.503 vs. odds ratio, 1.692 and
95% confidence interval, 1.371–2.087).
r P-value r P-value

Age, years 0.004 0.840 0.107 0.000 Correlation of SUA level with histological
BMI, kg/m2 0.218 0.000 0.059 0.007 severity in non-obese adults
FPG, mmol/L 0.061 0.006 0.098 0.000 To investigate the relationship between SUA level and dif-
TG, mmol/L 0.298 0.000 0.278 0.000
ferent stages of NAFLD in non-obese adults, 93 NAFLD pa-
TC, mmol/L 0.138 0.000 0.130 0.000
tients were divided into three groups according to biopsy
HDL-C, mmol/L 0.273 0.000 0.238 0.000
LDL-C, mmol/L 0.203 0.000 0.095 0.000
result (Table 5). An analysis using one-way ANOVA followed
ALT, U/L 0.185 0.000 0.129 0.000 by Dunnett’s test revealed that none of the variables had sig-
AST, U/L 0.093 0.000 0.186 0.000 nificant differences between non-NASH and NASH groups,
BUN, mmol/L 0.224 0.000 0.176 0.000 except for ALT and SUA. Notably, compared with the non-
SCr, mmol/L 0.421 0.000 0.320 0.000 NASH group, SUA levels were significantly higher in indefi-
nite NASH and NASH groups (344.53 ± 87.62 μmol/L vs.
Significant values appear in boldface type.
ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI,
397.71 ± 92.53 μmol/L and 403.21 ± 74.98 μmol/L, respec-
body mass index; BUN, blood urea nitrogen; FPG, fasting plasma tively). These results showed that non-obese adults had
glucose; HDL-C, high-density lipoprotein cholesterol; LDL-C, low- higher SUA levels as the inflammation advanced in NAFLD.
density lipoprotein cholesterol; r, correlation coefficient; SCr, serum
creatinine; TC, total cholesterol; TG, triglyceride.
Relationship between SUA level and
histopathological characteristics of NAFLD in
Risk factors analysis for NAFLD in non-obese non-obese adults
adults
The score distribution of NAS and its components accord-
Multivariate logistic regression analysis was undertaken to ing to SUA level is shown in Figure 2. The group with hy-
evaluate risk factors for NAFLD. Ten variables including peruricemia had significantly higher NAS scores and
male gender, age, TG, TC, HDL-C, LDL-C, FPG, SUA, more serious lobule inflammation than the normal group,
SCr, and BUN were entered into the original equation. whereas hepatic steatosis and ballooning degeneration
Our results showed that all variables except for male gen- had no significant difference between the two groups.
der, TC, SCr, and BUN remained in the final equation in To get a deeper understanding of the relationship be-
non-obese subjects. In obese subjects, male gender, age, tween SUA level and histopathological characteristics of
TG, FPG, and SUA remained in the final equation NAFLD patients in the non-obese group, the impact of
(Table 4). It was suggested that older age, hypertriglyc- SUA level on the prevalence rates of different NAS scores
eridemia, low HDL-C, elevated LDL-C and FPG, and hy- and its components were studied. Our results showed that
peruricemia are closely associated with the risk for the prevalence rates of high NAS and lobule inflammation

Table 4 Risk factors for non-alcoholic fatty liver disease in non-obese and obese adults

Variables Non-obese subjects Obese subjects

β P-value OR 95% CI β P-value OR 95% CI

Male gender – – – – 0.294 0.012 1.341 1.066–1.687

Older age 1.113 0.000 3.613 2.714–4.150 0.431 0.000 1.539 1.270–1.865
Elevated FPG 0.684 0.003 1.983 1.269–3.099 0.656 0.000 1.928 1.454–2.555
Hypertriglyceridemia 1.285 0.000 3.613 2.714–4.810 0.779 0.000 2.179 1.807–2.626
Low HDL-C 0.680 0.051 1.974 0.998–3.906 – – – –
Elevated LDL-C 0.684 0.000 1.981 1.483–2.648 – – – –
Hyperuricemia 0.940 0.000 2.559 1.870–3.503 0.526 0.000 1.692 1.371–2.087

–, none data; β, partial regression coefficient; CI, confidence interval; FPG, fasting plasma glucose; HDL-C, high-density lipoprotein cholesterol;
LDL-C, low-density lipoprotein cholesterol; OR, odds ratio.

© 2016 The Japan Society of Hepatology

Hepatology Research 2017; 47: E104–E112 Serum uric acid level E109

Table 5 Comparison of serum uric acid (SUA) level and other characteristics between non-alcoholic steatohepatitis (NASH), indefinite
NASH, and non-NASH in non-obese adults

Variable Total Non-NASH Indefinite NASH NASH P-value

n (male / female) 93 (67/26) 19 (15/4) 41 (28/13) 33 (24/9) 0.689

Age, years 36.84 ± 11.08 37.58 ± 9.71 35.83 ± 11.54 37.79 ± 11.59 0.738
BMI, kg/m2 23.44 ± 1.24 23.22 ± 1.23 23.62 ± 1.11 23.31 ± 1.41 0.469
TG, mmol/L 1.71 (1.27–2.79) 1.33(1.16–2.72) 1.62 (1.25–2.67) 1.94 (1.56–2.98) 0.307†
TC, mmol/L 5.06 ± 1.33 4.76 ± 0.99 5.08 ± 1.65 5.20 ± 1.01 0.324
HDL-C, mmol/L 1.27 ± 0.33 1.34 ± 0.29 1.20 ± 0.22 1.31 ± 0.43 0.193
LDL-C, mmol/L 3.08 ± 0.76 2.84 ± 0.70 3.25 ± 0.81 3.01 ± 0.69 0.295
FPG, mmol/L 5.26 ± 0.84 5.35 ± 1.15 5.20 ± 0.67 5.29 ± 0.82 0.835
Insulin, mU/mL 10.30 (8.77–15.30) 11.14 (7.38–18.19) 9.61 (8.77–15.30) 10.30 (8.00–12.30) 0.713†
HOMA-IR 2.20 (1.60–3.15) 2.30 (1.55–3.95) 2.05 (1.48–3.18) 2.30 (1.70–3.10) 0.615†
TNF-α, nmol/L 12.35 (8.48–17.85) 13.70 (9.70–17.02) 11.55 (6.00–18.24) 12.80 (8.85–19.55) 0.976†
AST, U/L 45.00 (30.00–69.5) 38.00 (24.00–55.00) 44.00 (28.00–62.50) 55.00 (33.00–80.50) 0.056†
ALT, U/L 75.00(42.00–116.50) 46.00 (28.00–98.00) 70.00 (44.00–96.5)‡ 100.00 (59.25–139.00)§ 0.009†
SUA, μmol/L 388.80 ± 87.72 344.53 ± 87.62 397.71 ± 92.53‡ 403.21 ± 74.98§ 0.030
Leptin, ng/mL 4.80 (2.60–9.30) 4.83 (2.53–15.6) 4.80 (2.68–9.30) 3.27 (1.90–10.00) 0.775†

Data are expressed as mean ± standard deviation or median (interquartile range). P-values calculated using one-way ANOVA, based on post hoc
multiple comparison between quartile groups.
†P-value was calculated using the Mann–Whitney U-test.
‡Compared with non-NASH, P < 0.05.
§Compared with indefinite NASH, P < 0.05.
ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; FPG, fasting plasma glucose; HDL-C, high-density li-
poprotein cholesterol; HOMA-IR, homeostasis model assessment of insulin resistance; LDL-C, low-density lipoprotein cholesterol; TC, total
cholesterol; TG, triglyceride; TNF-α, tumor necrosis factor-α.

score tended to increase as the SUA level increased,
whereas this trend was not apparent in hepatic steatosis
and ballooning degeneration (Fig. 3). In Q4, the preva-
lence rates of NAS ≥ 5 reached 57.58%, and the prevalence
rates of lobule inflammation score = 3 was 66.67% (Fig. 3).
It suggested that the inflammation of NAFLD tended to in-
crease as the SUA level increased in non-obese subjects.
DISCUSSION
increased SUA level is associated with an exacerbated risk
of NAFLD. This increased risk is probably independent of
conventional NAFLD risk factors.21 In our study, we also
found hyperuricemia was an risk factor of NAFLD. Non-
obese adults have higher NAFLD risk with increased SUA
levels than obese adults, suggesting that increased SUA
concentrations, even within the normal range, are associ-
ated with the presence of NAFLD in non-obese subjects.
Therefore, it is time to pay attention to the exact relation-
ship between SUA level and NAFLD in non-obese adults.

R ECENTLY, AN INCREASING number of studies have

examined the significance of NAFLD in non-obese
adults as NAFLD is increasingly prevalent in this group,
with reported prevalence from 11.5% to 16.1%.18,19 In this
study, we found that the prevalence of NAFLD was 14.88%
in non-obese adults. The NAFLD patients had higher SUA
levels than the control group. Previous studies showed that
hyperuricemia is a significant and independent predictor
of MetS, and MetS risk increases with increased serum uric
acid levels.20 Interestingly, the present study showed that
SUA levels were significantly and positively correlated with
BMI, TG, TC, LDL-C, ALT, SCr, and BUN, and the correla-
tion coefficient in the non-obese group was higher than
in the obese group. A meta-analysis suggested that
Large clinical studies have indicated that hyperuricemia
is associated with obesity, dyslipidemia, type 2 diabetes
mellitus, and MetS.22 There are also several studies suggest-
ing that SUA level is significantly associated with
NAFLD.12,13 However, few data are available about the re-
lationship between SUA level and inflammation associ-
ated with NAFLD in non-obese adults. In 93 cases of
non-obese biopsy-proven NAFLD we found that there
were no significant differences between non-NASH and
NASH groups, except for ALT and SUA levels. The SUA
levels were significantly higher in indefinite NASH and
NASH groups. These results showed that higher SUA
levels correlated with increased inflammation in NAFLD.
One of the possible explanations for the relationship

© 2016 The Japan Society of Hepatology

E110 J. Liu et al. Hepatology Research 2017; 47: E104–E112

Figure 2 Score distribution of non-alcoholic fatty liver disease activity score (NAS) and its components according to serum uric acid (SUA)
level. (a) NAS. (b) Hepatic steatosis. (c) Ballooning degeneration. (d) Lobule inflammation. P-values calculated using the Mann–Whitney U-test.

Figure 3 Prevalence of different quartile (Q) levels of serum uric acid according to the non-alcoholic fatty liver disease activity score
(NAS) and its components. (a) NAS. (b) Hepatic steatosis. (c) Ballooning degeneration. (d) Lobule inflammation. P-values calculated
using one-way ANOVA.*P < 0.05 versus Q1; †P < 0.05 versus Q2; ‡P < 0.05 versus Q3, based on post hoc multiple comparison between
quartile groups. , Q1; , Q2; , Q3; , Q4.

© 2016 The Japan Society of Hepatology

Hepatology Research 2017; 47: E104–E112
between SUA level and the inflammation progression
of NAFLD is that increased SUA is also linked to
increased fat accumulation, hyperleptinemia, and insulin
resistance.11,23
Uric acid becomes a strong oxidant in the environment of
MetS. Serum uric acid could also stimulate the release of
oxidants to promote oxidative stress by stimulating nicotin-
amide adenine dinucleotide phosphate oxidase and then be
involved in the pathogenesis of NAFLD. More interestingly,
in our study, patients with hyperuricemia had significantly
higher NAS scores than those without hyperuricemia, and
lobule inflammation was also more serious. The prevalence
rates of high NAS and lobule inflammation tended to in-
crease with increases in the SUA level. It suggested that the
inflammation of NAFLD tended to increase as the SUA level
increased. Recent studies discovered a close association be-
tween the inflammation development of NAFLD and many
pathogenic events, including hepatic macrophage recruit-
ment, activation of the innate immune system, and changes
in lipid homeostasis.24 The innate immune system has
evolved mechanisms to detect the release of a subset of
these molecules, called damage-associated molecular pat-
terns (DAMPs).25,26 When macrophages sense the presence
of DAMPs they are stimulated to produce pro-
inflammatory cytokines that then induce inflammation.27
It has been confirmed that uric acid is one of the important
DAMPs.28,29 Does uric acid act as a DAMP by stimulating
macrophages to produce pro-inflammatory cytokines and
then induce inflammation progression in non-obese
adults? Further research is needed.
In conclusion, the findings of the present study
suggest that hyperuricemia is an independent risk factor
for NAFLD, and SUA level is significantly associated
with inflammation progression of NAFLD in non-obese
adults. However, the precise mechanism requires further
study. Intervention trials are needed to investigate
whether the reduction of SUA levels favorably impacts
NAFLD.
ACKNOWLEDGMENTS
T HIS WORK WAS supported by the Natural Science
Foundation of Zhejiang Province (Grant Nos.
LY14H070004 and LR15H030001), the National Natural
Science Foundation of China (Grant Nos. 81100278 and
81470838), International Science and Technology Coop-
eration Projects of Zhejiang Province (Grant No.
2013C24010), Science Foundation of Health Bureau of
Zhejiang Province (Grant No. 2012RCA026 and
2013C33187), and Wang Baoen Liver Fibrosis Foundation
(Grant No. CFHPC20131041).
Serum uric acid level E111
REFERENCES
1 Malaguarnera M, Di Rosa M, Nicoletti F, Malaguarnera L. Mo-
lecular mechanisms involved in NAFLD progression. J Mol
Med (Berl) 2009; 87: 679–95.
2 Fan JG, Farrell GC. Epidemiology of non-alcoholic fatty liver
disease in China. J Hepatol 2009; 50: 204–10.
3 Xu C, Yu C, Ma H, Xu L, Miao M, Li Y. Prevalence and
risk factors for the development of nonalcoholic fatty
liver disease in a nonobese Chinese population: the
Zhejiang Zhenhai Study. Am J Gastroenterol 2013; 108:
1299–304.
4 Nishioji K, Sumida Y, Kamaguchi M et al. Prevalence of and
risk factors for non-alcoholic fatty liver disease in a non- obese
Japanese population, 2011–2012. J Gastroenterol 2015; 50:
95–108.
5 Edwards NL. The role of hyperuricemia in vascular disorders.
Curr Opin Rheumatol 2009; 21: 132–7.
6 Mellen PB, Bleyer AJ, Erlinger TP et al. Serum uric acid predicts
incident hypertension in a biethnic cohort: the atherosclerosis
risk in communities study. Hypertension 2006; 48: 1037–42.
7 Rho YH, Choi SJ, Lee YH et al. The prevalence of metabolic
syndrome in patients with gout: a multicenter study. J Korean
Med Sci 2005; 20: 1029–33.
8 Zhu Y, Pandya BJ, Choi HK. Comorbidities of gout and hyper-
uricemia in the US general population: NHANES 2007–2008.
Am J Med 2012; 125: 679–87.
9 Xu C. Hyperuricemia and nonalcoholic fatty liver disease:
from bedside to bench and back. Hepatol Int 2016; 10:
286–93.
10 Krishnan E, Pandya BJ, Chung L, Hariri A, Dabbous O. Hyper-
uricemia in young adults and risk of insulin resistance,
prediabetes, and diabetes: a 15-year follow-up study. Am J
Epidemiol 2012; 176: 108–16.
11 Yoo HG, Lee SI, Chae HJ et al. Prevalence of insulin resistance
and metabolic syndrome in patients with gouty arthritis.
Rheumatol Int 2011; 31: 485–91.
12 Li Y, Xu C, Yu C, Xu L, Miao M. Association of serum uric acid
level with non-alcoholic fatty liver disease: a cross-sectional
study. J Hepatol 2009; 50: 1029–34.
13 Sertoglu E, Ercin CN, Celebi G et al. The relationship of serum
uric acid with non-alcoholic fatty liver disease. Clin Biochem
2014; 47: 383–8.
14 Targher G, Bertolini L, Poli F et al. Nonalcoholic fatty liver dis-
ease and risk of future cardiovascular events among type 2
diabetic patients. Diabetes 2005; 54: 3541–6.
15 Kleiner DE, Brunt EM, Van Natta M et al. Design and valida-
tion of a histological scoring system for nonalcoholic fatty
liver disease. Hepatology 2005; 41: 1313–21.
16 Cerda C, Pérez-Ayuso RM, Riquelme A et al. Nonalcoholic
fatty liver disease in women with polycystic ovary syndrome.
J Hepatol 2007; 47: 412–7.
17 Bedogni G, Miglioli L, Masutti F, Tiribelli C, Marchesini G,
Bellentani S. Prevalence of and risk factors for nonalcoholic
fatty liver disease: the Dionysos nutrition and liver study.
Hepatology 2005; 42: 44–52.
© 2016 The Japan Society of Hepatology
E112 J. Liu et al.
18 Chen CH, Huang MH, Yang JC et al. Prevalence and risk fac-
tors of nonalcoholic fatty liver disease in an adult
population of Taiwan: metabolic significance of nonalcoholic
fatty liver disease in nonobese adults. J Clin Gastroenterol
2006; 40: 745–52.
19 Kim HJ, Kim HJ, Lee KE et al. Metabolic significance of nonal-
coholic fatty liver disease in nonobese, nondiabetic adults.
Arch Intern Med 2004; 164: 2169–75.
20 Nagahama K, Inoue T, Kohagura K, Ishihara A, Kinjo K, Ohya
Y. Hyperuricemia predicts future metabolic syndrome: a 4-
year follow-up study of a large screened cohort in Okinawa.
Japan Hypertens Res 2014; 37: 232–8.
21 Zhou Y, Wei F, Fan Y. High serum uric acid and risk of nonal-
coholic fatty liver disease: a syst- ematic review and meta-
analysis. Clin Biochem 2016; 49: 636–42.
22 Li C, Hsieh MC, Chang SJ. Metabolic syndrome, diabetes, and
hyperuricemia. Curr Opin Rheumatol 2013; 25: 210–6.
© 2016 The Japan Society of Hepatology
Hepatology Research 2017; 47: E104–E112
23 Rahmouni K, Haynes WG. Endothelial effects of leptin: impli-
cations in health and diseases. Curr Diab Rep 2005; 5: 260–6.
24 Baffy G. Kupffer cells in non-alcoholic fatty liver disease: the
emerging view. J Hepatol 2009; 51: 212–23.
25 Rock KL, Latz E, Ontiveros F, Kono H. The sterile inflamma-
tory response. Annu Rev Immunol 2010; 28: 321–42.
26 Rock KL, Lai JJ, Kono H. Innate and adaptive immune re-
sponses to cell death. Immunol Rev 2011; 243: 191–205.
27 Kono H, Karmarkar D, Iwakura Y, Rock KL. Identification of
the cellular sensor that stimulates the inflammatory response
to sterile cell death. J Immunol 2010; 184: 4470–8.
28 Kono H, Chen CJ, Ontiveros F, Rock KL. Uric acid promotes
an acute inflammatory response to sterile cell death in mice.
J Clin Invest 2010; 120: 1939–49.
29 Shi Y, Evans JE, Rock KL. Molecular identification of a danger
signal that alerts the immune system to dying cells. Nature
2003; 425: 516–21.