ORIGINAL ARTICLE

Is lower uric acid level better? A combined cross-sectionaland longitudinal study in the elderly

Chang-Hsun Hsieh • Jiunn-Diann Lin •

Chung-Ze Wu • Chun-Hsien Hsu • Dee Pei •

Yao-Jen Liang • Yen-Lin Chen

Received: 20 November 2013 / Accepted: 31 January 2014

� Springer Science+Business Media New York 2014

Abstract The relationship between metabolic syndrome

(MetS) and uric acid (UA) has been explored in many

studies, but there is paucity of information on hypo-uricemia

and MetS. The current study aimed to elucidate the rela-

tionship between the lower end of UA and MetS in elderly.

Subjects aged C60 years who underwent routine health

checkups were enrolled, and 10,579 were eligible for ana-

lysis. A cross-sectional study was first performed to deter-

mine the correlation between MetS and UA. The subjects

were divided into two groups according to the lowest male

MetS prevalence in each UA level (UA B5 and[5 mg/dl).

A longitudinal study then excluded subjects with MetS at

baseline to validate the UA level of those with the lowest

incidence of MetS. In the prevalence of MetS in different

UA levels, there was a J-shaped curve in males but a linear

relationship in females. In males, waist circumference, sys-

tolic blood pressure, high-density lipoprotein, and triglyc-

eride were correlated with UA in univariate analysis, but

high-density lipoprotein became insignificant in multivariate

analysis in UA [5 mg/dl (UA-high). Only fasting plasma

glucose (FPG) was significantly related to UA B5 mg/dl

(UA-low). The UA level of 4.5–5 mg/dl had the lowest risk

of having MetS in odds ratio, log-rank test, and hazard ratio.

A J-shaped phenomenon exists between UA and MetS in

elderly men but not in elderly women and FPG seem to be

the key factor in this. Lower UA does not provide benefits in

elderly men. The new UA control strategy for the elderly

warrants further investigation.

Keywords Uric acid � Metabolic syndrome � Elderly �J-shape

Introduction

Cardiovascular disease (CVD) is one of the leading causes of

death worldwide. Recently, its incidence has been increasing,

perhaps due to longer life span and aging. For the early

detection of people at high risk of CVD, the concept of meta-

bolic syndrome (MetS) has been elaborated. First described by

C.-H. Hsieh

Division of Endocrinology and Metabolism, Department of

Internal Medicine, Tri-Service General Hospital, National

Defense Medical Center, Taipei, Taiwan

J.-D. Lin � C.-Z. Wu

Division of Endocrinology and Metabolism, Department of

Medicine, Shuang-Ho Hospital, Taipei Medical University,

New Taipei, Taiwan

C.-H. Hsu

Department of Family Medicine, Cardinal Tien Hospital, School

of Medicine, Fu-Jen Catholic University, New Taipei, Taiwan

D. Pei

Department of Internal Medicine, Cardinal Tien Hospital, School

of Medicine, Fu-Jen Catholic University, New Taipei, Taiwan

Y.-J. Liang

Department of Life-Science, Fu-Jen Catholic University,

New Taipei, Taiwan

Y.-L. Chen (&)

Department of Pathology, Cardinal Tien Hospital, School of

Medicine, Fu-Jen Catholic University, No. 362, Chung-Cheng

Road, Xindian, New Taipei 23137, Taiwan

e-mail: anthonypatho@gmail.com

123

Endocrine

DOI 10.1007/s12020-014-0201-6

Reaven et al. [1] in 1988, MetS is characterized by the clus-

tering of dyslipidemia, obesity, hypertension, and hypergly-

cemia. Subsequently, MetS has been highly associated with the

development of CVD and type 2 diabetes mellitus (T2D).

Initially, the diagnostic criteria of MetS were not the same in

studies across the globe and more than forty criteria were

proposed [2]. The National Cholesterol Education Program

Adult Treatment Panel III eventually defined the most popular

diagnostic criteria of MetS [3], which contained hyperglyce-

mia, central obesity, hypertension, and dyslipidemia. However,

these components are the so-called ‘‘traditional criteria’’. Other

risk factors such as uric acid (UA) or increased liver enzyme

have been discovered and proven to be related to CVD. These

are regarded as the ‘‘non-traditional’’ risk factors for MetS [4–

7]. The prevalence of MetS in Taiwan is around 20 % among

all age of population. However, the rate is much higher in

elderly with more than half of the elderly has MetS [8].

A by-product of purine metabolism generated during the

breakdown of nucleic acids, UA at higher level is related to

hypertension and insulin resistance [9]. However, findings

on the relationships between UA and T2D are inconsistent

[10–15]. Bo et al. [16] demonstrated that hypo-uricemia and

hyper-uricemia seem to be two different phenotypes in T2D.

Hyper-uricemia was associated with the insulin-resistant

syndrome and early onset or increased progression to overt

nephropathy, while hypo-uricemia was associated with

worse metabolic control, hyper-filtration, and late onset or

decreased progression to overt nephropathy. In another study

by Yuan et al. [17], UA level was initially increased and then

decreased as glucose levels increased from normoglycemia

due to impaired glucose regulation and T2D. Although the

relationship between MetS and UA, especially hyper-urice-

mia, has been the subject of many studies, there is a lack of

information on hypo-uricemia and MetS. The current cohort

study tried to shed light on the relationship between the lower

end of UA and MetS in elderly adults.

Methods

Study population

Individuals aged C60 years (included) who underwent

routine health checkups at the MJ Health Screening Center,

a privately owned chain of clinics throughout Taiwan that

provided regular annual health examinations to their

members, were enrolled. The nationwide distribution of the

MJ Health Screening Centers provided a selection sample

from across the entire country. However, the study par-

ticipants had higher socioeconomic status than the general

population, which would be a bias of selection. All of the

study participants were anonymous and provided informed

consent. Data were provided by the MJ Health Screening

Center for research purposes only and the center’s insti-

tutional review board approved the study protocol.

From the database covering the period 1999 and 2007,

21,637 records were randomly selected. Initially, 2,418

subjects were excluded due to only a single visit to the

center. A more stringent exclusion criteria to make the

results more reliable were used by reducing the con-

founding effects of medications for hypertension, hyper-

lipidemia and/or T2D at baseline, thereby excluding

another 8,423 subjects. After excluding 217 subjects with

missing data of MetS components, a total of 10,579 were

eligible for further analysis (‘‘Appendix 1’’).

Anthropometric measurements and general data

Members of the senior nursing staff used a questionnaire to

obtain the subjects’ medical history, including current medi-

cations. Physicians then performed a thorough history taking

and complete physical examination. Body mass index (BMI)

was calculated as the subject’s body weight (kg) divided by

the square of the subject’s height (m). Waist circumference

(WC) was taken at the midway point between the inferior

margin of the last rib and the crest of the ilium in a horizontal

plane. Both systolic blood pressure (SBP) and diastolic blood

pressure (DBP) were measured by senior nursing staff using

mercury sphygmomanometers with appropriate-sized cuffs

on the right arm of the participants, who had rested for at least

5 min in a sitting position. Two measurements were taken

more than 1 min apart and the average was recorded. Blood

samples were drawn from the antecubital vein after 10-h

fasting for laboratory and biochemical analysis. Plasma was

separated from blood within 1 h, stored at -70 �C and ana-

lyzed for fasting plasma glucose (FPG) and lipid profiles. The

FPG was detected using the glucose oxidase method (YSI 203

glucose analyzer, Scientific Division, Yellow Springs Instru-

ments, Yellow Springs, OH). Total cholesterol and triglycer-

ides (TG) were measured using the dry, multi-layer analytical

slide method in a Fuji Dri-Chem 3000 analyzer (Fuji Photo

Film, Minato-Ku, Tokyo, Japan). Serum high-density lipo-

protein (HDL) and low-density lipoprotein (LDL) cholesterol

concentrations were analyzed using an enzymatic cholesterol

assay following dextran sulfate precipitation. Serum UA was

measured by the uricase-peroxidase method using an Abbott

C8000 analyzer (Abbott Laboratories, Abbott Park, IL, USA).

Definition of metabolic syndrome

The latest harmonized criteria of MetS in 2009 [18] were

used, with some modifications. The WC was C90 and

80 cm for Taiwanese males and females, respectively [19].

Other four criteria were the same: SBP C 130 mmHg or

DBP C 85 mmHg, TG C 150 mg/dl, FPG C 100 mg/dl,

HDL B 40 and 50 mg/dl for males and females,

Endocrine

123

respectively, or intake of related medications. Subjects had

to have at least three criteria to be diagnosed as MetS.

Statistical analysis

The study had two parts. The first part was a cross-sectional

study aimed to determine the association between UA and

MetS. The subjects were divided into two groups according to

the lowest male MetS prevalence in each UA level (UA-high

or UA level [5 mg/dl and UA-low, or UA level B5 mg/dl).

T test was used to compare differences in demographic data

between these two groups. Pearson’s simple correlation was

applied to identify the MetS criterion related to UA. Multiple

regression analysis with adjustment was then performed to

differentiate components that were independently correlated

with UA. Odds ratio (OR) by logistic regression analysis was

also applied for the risk of having MetS.

The second part of the study had a longitudinal design to

demonstrate the cause–effect relationship of UA and MetS.

Kaplan–Meier plot with log-rank test was done to show the

proportion of MetS-free subjects during the follow-up

period. Lastly, Cox regression was used to see the hazard

ratio (HR) of developing MetS.

All statistical analyses were performed using the SPSS

18.0 software (SPSS Inc., Chicago, IL). Data were pre-

sented as mean ± standard deviation. All data were tested

by the Kolmogorov–Smirnov test for normal distribution

and by the Levene’s test for homogeneity of variances.

Statistical significance was set at a two-sided p \ 0.05.

Results

The study enrolled 10,579 elderly in the current study.

Based on the prevalence of MetS in different levels of UA

(each 0.5 mg/dl as a subgroup) (Fig. 1), there was a

J-shaped curve in males and a linear relationship in

females. The lowest prevalence of MetS was in the group

of UA level 4.5–5 mg/dl in males. The study population

was divided into two groups (UA-high and UA-low) based

on the prevalence result. On the demographic data

(Table 1), except for age, SBP in males and FPG in

females, all other MetS components (WC, BP, HDL, and

TG) were higher in the UA-high group.

In males, all MetS components except FPG were cor-

related with UA in the univariate analysis, but DBP andFig. 1 The prevalence of metabolic syndrome according to uric acid

level

Table 1 Demographic data of study subjects with uric acid of 5 mg/

dl as cut-off point

UA-low UA-high p

UA B5

(mg/dl)

UA [5

(mg/dl)

Male

n 630 4,509

Age (years) 65.3 ± 5.5 64.8 ± 5.2 0.054

Body mass index (kg/m2) 22.64 ± 2.90 23.73 ± 2.86 \0.001

Waist circumference (cm) 81.3 ± 8.7 84.2 ± 8.4 \0.001

Systolic blood pressure

(mmHg)

127.5 ± 18.5 128.7 ± 18.4 0.134

Diastolic blood pressure

(mmHg)

74.5 ± 10.8 76.2 ± 11.1 \0.001

Fasting Plasma Glucose

(mg/dl)

108.7 ± 37.7 102.0 ± 16.7 \0.001

Total cholesterol (mg/dl) 196.7 ± 34.7 201.8 ± 35.2 0.001

HDL-C (mg/dl) 52.2 ± 13.9 49.8 ± 13.4 \0.001

LDL-C (mg/dl) 123.4 ± 30.4 127.1 ± 31.8 0.005

Triglyceride (mg/dl) 105.6 ± 55.9 124.5 ± 64.5 \0.001

Log TG 1.98 ± 0.20 2.04 ± 0.21 \0.001

Uric acid (mg/dl) 4.44 ± 0.54 6.94 ± 1.28 \0.001

Female

n 2,173 3,267

Age (years) 63.7 ± 4.4 63.7 ± 4.4 0.915

Body mass index (kg/m2) 22.96 ± 3.02 24.47 ± 3.16 \0.001

Waist circumference (cm) 75.1 ± 7.6 78.4 ± 7.9 \0.001

Systolic blood pressure

(mmHg)

128.7 ± 18.4 131.3 ± 19.1 \0.001

Diastolic blood pressure

(mmHg)

72.6 ± 10.6 74.6 ± 10.8 \0.001

Fasting plasma glucose

(mg/dl)

100.8 ± 22.5 101.0 ± 17.2 0.772

Total cholesterol (mg/dl) 211.4 ± 36.1 216.8 ± 37.5 \0.001

HDL-C (mg/dl) 62.0 ± 15.6 57.4 ± 14.9 \0.001

LDL-C (mg/dl) 127.7 ± 32.6 132.7 ± 34.4 \0.001

Triglyceride (mg/dl) 107.4 ± 51.6 133.7 ± 66.4 \0.001

Log TG 1.99 ± 0.19 2.08 ± 0.20 \0.001

Uric acid (mg/dl) 4.31 ± 0.55 6.31 ± 1.08 \0.001

Data are shown as mean ± SEM

UA uric acid

Endocrine

123

HDL became insignificant in the multivariate analysis in

the UA-high group (‘‘Appendix 2’’). In contrast, only FPG

was significantly correlated with the UA-low group. In

females, only WC and TG remain significantly correlated

with UA in the multivariate analysis regardless of UA level

(both UA-low and UA-high). Using OR, a UA level of

4.5–5 and 3.5–4 mg/dl had the lowest risk of having MetS

in males (p \ 0.01) and females, respectively (Fig. 2).

Compared the lowest UA level (3.5–4 mg/dl) with

4.5–5 mg/dl in men, OR of 1.920 (1.054–3.496) was seen

for having MetS.

In second part of the study, the median follow-up period

was 5.6 years. After excluding 2,861 subjects with MetS at

baseline, a total of 7,718 subjects were eligible for follow-

up analysis not only for the cause–effect relationships

between UA and MetS, but also as validation to confirm the

result in first part of study. The Kaplan–Meier plot showed

similar results that the group with UA level of 4.5–5 mg/dl

would have the lowest risk of developing MetS (‘‘Appen-

dix 3’’). In addition, the Cox regression model (Fig. 3)

further confirmed the results, with a HR of 1.972

(1.091–3.564) when the UA level was 3.5–4 mg/dl (lowest)

compared to that when the UA level was 4.5–5 mg/dl

(p = 0.025). Instead of increase MetS risk in the lowest

UA level (3.5–4 mg/dl) in male, it still has the lowest risk

for having MetS in both Kaplan–Meier plot and HR for

having MetS in female. Thus, both parts of the study

revealed an inverse correlation between hypo-uricemia and

MetS only in elderly males.

Discussion

To date, this study is the first to describe the J-shaped

relationship between UA and MetS in elderly men. In

contrast, elderly women have a linear relationship between

these two factors. In the prevalence of MetS but also in the

OR and HR of having MetS, the UA level of 4.5–5 mg/dl

has the lowest risk of having MetS in elderly men. By

Fig. 2 Odds ratio of having metabolic syndrome in different levels of

uric acid. (Asterisk) Uric acid level of 4.5–5 mg/dl was the reference rangeFig. 3 Hazard ratio of having metabolic syndrome in different levels

of uric acid during the follow-up period. (Asterisk) Uric acid level of

4.5–5 mg/dl was the reference range

Endocrine

123

univariate and multivariate analyses, FPG seems to play the

most critical role. However, this phenomenon is not seen in

elderly women. Although Denzer et al. [20] have found

that testosterone is positively correlated with serum UA in

obese children and adolescents, these results suggest

another example of an undescribed impact of androgens in

the regulation of UA.

The relationship between UA and FPG is a controversial

issue. Some authors have found no association between the

two, others reveal an inverse relationship, and still another

study shows a positive relationship [21–23]. Moreover, a

bell-shaped relationship between UA and glucose levels

has also been reported [24, 25]. Chou et al. [26] found that

hyper-uricemia and HOMA-insulin resistance were posi-

tively associated with hyper-insulinemia among subjects

without T2D in both sexes. The association among UA,

insulin resistance, and plasma glucose levels was stronger

in females than in males. In the study done by Robles-

Cervantes et al. [27], a positive association between UA

and insulin secretion (first, second, and total phases) was

noted in T2D male patients but not in females by the

hyperglycemic clamp technique.

In addition, Meisinger et al. [28] separated the study

group into different categories according to FPG level.

They found that UA was associated with isolated impaired

fasting glucose, impaired fasting glucose, impaired glucose

tolerance, newly diagnosed T2D, and known T2D in

women. However, only isolated impaired fasting glucose

was significantly related to UA in men. A possible mech-

anism may be that UA inhibits glucose-induced insulin

secretion by binding to an essential arginine residue in rat

pancreatic beta-cells [29]. Results of the present study

show that only FPG is significantly negatively correlated

with UA among the factors in males of the UA-low group.

This can explain the J-shape phenomenon in elderly men.

Waist circumference has positively correlated with UA

and discussed extensively. Sui et al. demonstrated that UA

was positively related to WC in 9,689 subjects aged

20–82 years [30]. At the same time, Liu et al. [22] also

showed similar results in another Taiwanese cohort.

Results here further confirmed this relationship in the

elderly regardless of sexes. However, there was an insig-

nificant correlation between WC and UA in the UA-low

group of elderly men. As mentioned previously, Denzer

et al. [20] have that testosterone positively correlated with

serum UA in obese children and adolescents. Although age

might be different in these situations, Marinello et al. [31]

further found that the purine nucleotide metabolism slowed

down after castration and was not completely restored by

androgen administration. These may explain the sex dif-

ference but not the different scenarios of UA-high and

UA-low. The underlying mechanisms warrant further

investigation.

Studies have been conducted in the field of hypertension

and UA. It has been reported that hyper-uricemia carries an

increased risk of hypertension [32]. In healthy subjects,

increased UA level, even within the physiologic range

(within normal limits), is a risk factor for increased carotid

intima-media thickness, reduced brachial artery flow-

mediated dilatation, and increased aortic stiffness inde-

pendent of other cardiovascular risk factors or other factors

related to MetS [33]. However, in the present study, only

SBP is significantly correlated with elderly men in the UA-

high group. The major difference between the present study

and a previous study is the age of the study population.

Sundstrom et al. [34] also failed to find the association

between UA and hypertension in the elderly. Increasing

age is associated with stiffening aorta, activated renin-

angiotensin system, and renal vasoconstriction. These may

blunt the association with hypertension. [35, 36].

Although HDL is negatively correlated with UA as

expected in univariate regression among women in both the

UA-high and UA-low groups and in men in the UA-high

group, this becomes insignificant in multivariate analysis.

In contrast, TG is still significantly related to UA except

among males in the UA-low group. Ishizaka et al. [37] also

found that UA was associated with dyslipidemia in Japa-

nese individuals. This indirectly supports the present find-

ings. However, Ishizaka et al. did not separate UA from

5 mg/dl (UA-high and UA-low) and still showed a positive

linear relationship between TG and UA. It is known that

dyslipidemia and high TG are indicative of pro-inflam-

matory state/oxidative stress [38]. Several studies suggest

that UA also has pro-inflammatory effects [39, 40].

Moreover, pro-inflammatory state/oxidative stresses often

precede or are prominently involved in MetS [41]. There-

fore, UA may contribute to the development of MetS

through a pro-inflammatory pathway.

The strength of this study is its combined cross-sectional

and longitudinal aspects. This further supports and

strengthens the findings of an inverse relationship between

MetS and hypo-uricemia in elderly men. Nonetheless, there

are still some limitations in this study. First, the subjects

were recruited from one private health screening center.

Thus, they had better economic status with more medical

support and may not represent the conditions of the general

population in Taiwan. Second, subjects taking medications

related to MetS components were excluded. Although this

may reduce the confounding factors and provides a clearer

relationship between MetS and UA, this condition is not

often seen in general clinical practice. As such, the strict

exclusion criteria are both an advantage and a disadvantage

in the current study. Although the UA level of the lowest

risk having MetS has been determined in a cross-sectional

study, the longitudinal cohort study may be seen as a

validation. However, no further external validation was

Endocrine

123

done, which may be another weakness of the current study.

Lastly, some of the study population had several risk fac-

tors including hypertension and dyslipidemia. Although we

excluded subjects taking medication would affect meta-

bolic syndrome at baseline to reduce the effect of medi-

cation, it is still impossible for us to make the strong

evidence to eliminate the possible effect of underlying

diseases and other medications.

In conclusion, a J-shape phenomenon exists between

UA and MetS in elderly men but not in elderly women and

FPG seem to be a key factor. Lower UA does not provide

more benefits for elderly men. New treatment plans for UA

control in the elderly are needed and warrant further

investigations.

Acknowledgments The authors sincerely thank the MJ Life Clinic

in Taiwan for providing the anonymous data for the present study.

Conflict of interest All the authors have no conflicts of interest in

the present study.

Appendix 1

See Fig. 4.

Fig. 4 Description of the study

design

Endocrine

123

Appendix 2

See Table 2.

Table 2 The Kaplan–Meier estimates of metabolic syndrome during the follow-up period according to the uric acid level at baseline

Univariant Multivariant

UA-low UA-high UA-low UA-high

UA B5 UA [5 UA B5 UA [5

Male n = 630 n = 4,509 n = 630 n = 4,509

Age r -0.013 -0.009 -0.001 -0.047

p 0.749 0.544 0.980 0.002

Waist circumference r -0.004 0.189 0.113 0.129

p 0.927 \0.001 0.150 \0.001

Systolic blood pressure r -0.079 0.090 -0.072 0.052

p 0.046 \0.001 0.229 0.015

Diastolic blood pressure r -0.058 0.083 0.011 0.003

p 0.148 \0.001 0.851 0.869

Fasting plasma glucose r -0.130 -0.003 -0.127 -0.050

p 0.001 0.834 0.002 \0.001

High-density

lipoprotein

r 0.033 -0.102 0.012 -0.009

p 0.407 \0.001 0.786 0.553

Triglyceride r -0.064 0.174 -0.042 0.112

p 0.111 \0.001 0.350 \0.001

Female n = 2,173 n = 3,267 n = 2,173 n = 3,267

Age r -0.015 0.050 -0.015 0.017

p 0.488 0.005 0.508 0.361

Waist circumference r 0.086 0.165 0.023 0.066

p \0.001 \0.001 0.546 0.023

Systolic blood pressure r -0.005 0.032 -0.048 0.007

p 0.810 0.071 0.127 0.776

Diastolic blood pressure r 0.025 0.020 0.030 -0.029

p 0.250 0.256 0.317 0.245

Fasting plasma glucose r -0.011 0.019 -0.035 -0.014

p 0.594 0.283 0.105 0.414

High-density

lipoprotein

r -0.065 -0.107 -0.014 -0.029

p 0.002 \0.001 0.554 0.128

Triglyceride r 0.098 0.131 0.072 0.093

p \0.001 \0.001 0.003 \0.001

All p value was adjusted for smoking status, alcohol consumption, BUN, creatinin, body mass index, and LDL-cholesterol in multivariate

regression analysis

Unit = mg/dl

UA uric acid

*Log-rank test for p \ 0.05 compared to the reference range of uric acid. #Uric acid level of 4.5–5 mg/dl was the reference range

Endocrine

123

Appendix 3

See Fig. 5.

Fig. 5 Univariate and

multivariate analyses of uric

acid and components of

metabolic syndrome

Endocrine

123

References

1. G.M. Reaven, Banting lecture 1988. Role of insulin resistance in

human disease. Diabetes 37(12), 1595–1607 (1988)

2. N.M. Varda, A. Gregoric, Metabolic syndrome in the pediatric

population: a short overview. Pediatric Rep 1(1), e1 (2009)

3. N.J. Stone, S. Bilek, S. Rosenbaum, Recent National Cholesterol

Education Program Adult Treatment Panel III update: adjust-

ments and options. Am. J. Cardiol. 96(4A), 53E–59E (2005)

4. C.H. Hsu, J.Y. Wang, Y.L. Chen, C.C. Liu, Y.L. Chang, H.S. Chen,

C. Pei, D. Pei, Relationships between alanine aminotransferase

levels, abnormal liver echogenicity, and metabolic syndrome. J. Am.

Board Fam. Med. JABFM 24(4), 407–414 (2011)

5. J.Y. Wang, Y.L. Chen, C.H. Hsu, S.H. Tang, C.Z. Wu, D. Pei,

Predictive value of serum uric acid levels for the diagnosis of met-

abolic syndrome in adolescents. J. Pediatr. 161(4), 753–756 (2012)

6. J.P. Goncalves, A. Oliveira, M. Severo, A.C. Santos, C. Lopes,

Cross-sectional and longitudinal associations between serum uric

acid and metabolic syndrome. Endocrine 41(3), 450–457 (2012).

doi:10.1007/s12020-012-9629-8

7. R.D. Santos, Elevated uric acid, the metabolic syndrome and

cardiovascular disease: cause, consequence, or just a not so

innocent bystander? Endocrine 41(3), 350–352 (2012)

8. National Health Research Institutes. http://nahsit.nhri.org.tw/

node/21. Accessed January 10

9. K. Masuo, H. Kawaguchi, H. Mikami, T. Ogihara, M.L. Tuck,

Serum uric acid and plasma norepinephrine concentrations pre-

dict subsequent weight gain and blood pressure elevation.

Hypertension 42(4), 474–480 (2003)

10. A. Dehghan, M. van Hoek, E.J. Sijbrands, A. Hofman, J.C.

Witteman, High serum uric acid as a novel risk factor for type 2

diabetes. Diabetes Care 31(2), 361–362 (2008)

11. C.K. Kramer, D. von Muhlen, S.K. Jassal, E. Barrett-Connor,

Serum uric acid levels improve prediction of incident type 2

diabetes in individuals with impaired fasting glucose: the Rancho

Bernardo Study. Diabetes Care 32(7), 1272–1273 (2009)

12. S. Kodama, K. Saito, Y. Yachi, M. Asumi, A. Sugawara, K. Totsuka,

A. Saito, H. Sone, Association between serum uric acid and devel-

opment of type 2 diabetes. Diabetes Care 32(9), 1737–1742 (2009)

13. E. Oda, R. Kawai, V. Sukumaran, K. Watanabe, Uric acid is

positively associated with metabolic syndrome but negatively

associated with diabetes in Japanese men. Intern. Med. 48(20),

1785–1791 (2009)

14. H. Nan, Y. Dong, W. Gao, J. Tuomilehto, Q. Qiao, Diabetes

associated with a low serum uric acid level in a general Chinese

population. Diabetes Res. Clin. Pract. 76(1), 68–74 (2007)

15. P. Bandaru, A. Shankar, Association between serum uric acid levels

and diabetes mellitus. Int. J. Endocrinol. 2011, 604715 (2011)

16. S. Bo, P. Cavallo-Perin, L. Gentile, E. Repetti, G. Pagano,

Hypouricemia and hyperuricemia in type 2 diabetes: two differ-

ent phenotypes. Eur. J. Clin. Invest. 31(4), 318–321 (2001)

17. H.J. Yuan, X.G. Yang, X.Y. Shi, R. Tian, Z.G. Zhao, Association

of serum uric acid with different levels of glucose and related

factors. Chin. Med. J. 124(10), 1443–1448 (2011)

18. K.G. Alberti, R.H. Eckel, S.M. Grundy, P.Z. Zimmet, J.I. Cle-

eman, K.A. Donato, J.C. Fruchart, W.P. James, C.M. Loria, S.C.

Smith Jr, Harmonizing the metabolic syndrome: a joint interim

statement of the International Diabetes Federation Task Force on

Epidemiology and Prevention; National Heart, Lung, and Blood

Institute; American Heart Association; World Heart Federation;

International Atherosclerosis Society; and International Associ-

ation for the Study of Obesity. Circulation 120(16), 1640–1645

(2009)

19. Department of Health, Executive Yuan, R.O.C. (Taiwan).

Available at: http://www.doh.gov.tw/CHT2006/DM/DM2_p01.

aspx?class_no=25&now_fod_list_no=5912&level_no=2&doc_

no=22602. Accessed February 10, 2013

20. C. Denzer, R. Muche, H. Mayer, E. Heinze, K.M. Debatin, M.

Wabitsch, Serum uric acid levels in obese children and adoles-

cents: linkage to testosterone levels and pre-metabolic syndrome.

J. Pediatr. Endocrinol. Metab. JPEM 16(9), 1225–1232 (2003)

21. J.D. Lin, W.K. Chiou, H.Y. Chang, F.H. Liu, H.F. Weng, Serum

uric acid and leptin levels in metabolic syndrome: a quandary

over the role of uric acid. Metab. Clin. Exp. 56(6), 751–756

(2007)

22. P.W. Liu, T.Y. Chang, J.D. Chen, Serum uric acid and metabolic

syndrome in Taiwanese adults. Metab. Clin. Exp. 59(6), 802–807

(2010)

23. W.K. Chiou, D.H. Huang, M.H. Wang, Y.J. Lee, J.D. Lin, Sig-

nificance and association of serum uric acid (UA) levels with

components of metabolic syndrome (MS) in the elderly. Arch.

Gerontol. Geriatr. 55(3), 724–728 (2012)

24. D.G. Cook, A.G. Shaper, D.S. Thelle, T.P. Whitehead, Serum

uric acid, serum glucose and diabetes: relationships in a popu-

lation study. Postgrad. Med. J. 62(733), 1001–1006 (1986)

25. T.P. Whitehead, I. Jungner, D. Robinson, W. Kolar, A. Pearl, A.

Hale, Serum urate, serum glucose and diabetes. Ann. Clin. Bio-

chem. 29(Pt 2), 159–161 (1992)

26. P. Chou, K.C. Lin, H.Y. Lin, S.T. Tsai, Gender differences in the

relationships of serum uric acid with fasting serum insulin and

plasma glucose in patients without diabetes. J. Rheumatol. 28(3),

571–576 (2001)

27. J.A. Robles-Cervantes, M.G. Ramos-Zavala, M. Gonzalez-Ortiz,

E. Martinez-Abundis, C. Valencia-Sandoval, A. Torres-Chavez,

C. Espinel-Bermudez, N.J. Santiago-Hernandez, S.O. Hernandez-

Gonzalez, Relationship between serum concentration of uric acid

and insulin secretion among adults with type 2 diabetes mellitus.

Int. J. Endocrinol. 2011, 107904 (2011)

28. C. Meisinger, A. Doring, D. Stockl, B. Thorand, B. Kowall, W.

Rathmann, Uric acid is more strongly associated with impaired

glucose regulation in women than in men from the general

population: the KORA F4-Study. PLoS ONE 7(5), e37180 (2012)

29. B. Rocic, M. Vucic-Lovrencic, N. Poje, M. Poje, F. Bertuzzi,

Uric acid may inhibit glucose-induced insulin secretion via

binding to an essential arginine residue in rat pancreatic beta-

cells. Bioorg. Med. Chem. Lett. 15(4), 1181–1184 (2005)

30. X. Sui, T.S. Church, R.A. Meriwether, F. Lobelo, S.N. Blair, Uric

acid and the development of metabolic syndrome in women and

men. Metab. Clin. Exp. 57(6), 845–852 (2008)

31. E. Marinello, R. Leoncini, L. Terzuoli, D. Vannoni, B. Porcelli,

G. Resconi, Effect of testosterone on purine nucleotide metabo-

lism in rat liver. Hormone Metabol Res = Hormon- und

Stoffwechselforschung = Hormones et metabolisme 36(9), 614–

619 (2004)

32. D.I. Feig, D.H. Kang, R.J. Johnson, Uric acid and cardiovascular

risk. N. Engl. J. Med. 359(17), 1811–1821 (2008)

33. D. Erdogan, H. Gullu, M. Caliskan, E. Yildirim, M. Bilgi, T.

Ulus, N. Sezgin, H. Muderrisoglu, Relationship of serum uric

acid to measures of endothelial function and atherosclerosis in

healthy adults. Int. J. Clin. Pract. 59(11), 1276–1282 (2005)

34. J. Sundstrom, L. Sullivan, R.B. D’Agostino, D. Levy, W.B.

Kannel, R.S. Vasan, Relations of serum uric acid to longitudinal

blood pressure tracking and hypertension incidence. Hyperten-

sion 45(1), 28–33 (2005)

35. T.S. Perlstein, O. Gumieniak, P.N. Hopkins, L.J. Murphey, N.J.

Brown, G.H. Williams, N.K. Hollenberg, N.D. Fisher, Uric acid

and the state of the intrarenal renin-angiotensin system in

humans. Kidney Int. 66(4), 1465–1470 (2004)

36. S. Watanabe, D.H. Kang, L. Feng, T. Nakagawa, J. Kanellis, H.

Lan, M. Mazzali, R.J. Johnson, Uric acid, hominoid evolution,

Endocrine

123

and the pathogenesis of salt-sensitivity. Hypertension 40(3),

355–360 (2002)

37. N. Ishizaka, Y. Ishizaka, E. Toda, R. Nagai, M. Yamakado,

Association between serum uric acid, metabolic syndrome, and

carotid atherosclerosis in Japanese individuals. Arterioscler.

Thromb. Vasc. Biol. 25(5), 1038–1044 (2005)

38. R. Ross, Atherosclerosis—an inflammatory disease. N. Engl.

J. Med. 340(2), 115–126 (1999)

39. J. Kanellis, D.H. Kang, Uric acid as a mediator of endothelial

dysfunction, inflammation, and vascular disease. Semin. Nephrol.

25(1), 39–42 (2005)

40. A.C. Gagliardi, M.H. Miname, R.D. Santos, Uric acid: a marker

of increased cardiovascular risk. Atherosclerosis 202(1), 11–17

(2009)

41. A. Onat, Metabolic syndrome: nature, therapeutic solutions and

options. Expert Opin. Pharmacother. 12(12), 1887–1900 (2011)

Endocrine

123