Preventive Medicine 90 (2016) 201–206

Contents lists available at ScienceDirect

Preventive Medicine

journal homepage: www.elsevier.com/locate/ypmed

Serum uric acid as a predictor of future hypertension: Stratified analysis

based on body mass index and age
Yuki Yokoi a, Takahisa Kondo b,⁎, Naoki Okumura a, Keiko Shimokata a, Shigeki Osugi a,
Kengo Maeda a, Toyoaki Murohara a
a
Department of Cardiology, Nagoya University Graduate School of Medicine, Nagoya, Japan
b
Department of Advanced Medicine in Cardiopulmonary Disease, Nagoya University Graduate School of Medicine, Nagoya, Japan

a r t i c l e i n f o a b s t r a c t

Article history: Background. Serum uric acid level is a predictor of future hypertension. However, its dependence on body
Received 16 February 2016 mass index or age is unclear.
Received in revised form 1 June 2016 Methods. We examined 26,442 Japanese males aged 18–60 years free from hypertension or diagnosed cardio-
Accepted 8 July 2016
vascular disease at baseline followed up between 2000 and 2010. Participants were categorized into three groups
Available online 9 July 2016
according to the tertile of serum uric acid levels [mg/dL; 1st (reference): 0.1–5.3; 2nd: 5.4–6.2; 3rd: 6.3–11.6].
Keywords:
Incident hypertension was defined as newly detected blood pressure ≥ 140/90 mm Hg and/or antihypertensive
Blood pressure drugs initiation. Body mass index (b25 kg/m2 vs. ≥25 kg/m2) and age (b40 years vs. ≥40 years) were stratified
Epidemiology into two groups.
Hypertension Results. During a mean follow-up of 7.2 years, there were 11,361 (43%) hypertension cases. Mean serum uric
Uric acid acid levels (mg/dL) at baseline in each group were 1st tertile, 4.6; 2nd tertile, 5.8; and 3rd tertile, 7.0. The cumu-
lative incident hypertension rate was significantly higher in the 3rd tertile (50.8%) than in the 1st (37.4%). Mul-
tiple-adjusted hazard ratios (95% confidence interval) for incident hypertension compared with 1st tertile were
1.01 (0.96–1.05) and 1.15 (1.10–1.21) in the 2nd and 3rd tertile, respectively. There was a significant interaction
between age and serum uric acid level (p for interaction = 0.035). In subjects aged ≥40 years, the 3rd serum uric
acid group showed higher hazard ratios [1.48 (1.38–1.59)].
Conclusion. High serum uric acid level was associated with future hypertension in young and middle-aged
Japanese males. This association was stronger among subjects ≥40 years old.
© 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction 2007; Mellen et al., 2006; Nakanishi et al., 2003; Taniguchi et al., 2001).
Some studies have shown that blood pressure is lowered by UA-lowering
Hypertension is one of the most important risk factors of stroke, car- drugs (Agarwal et al., 2013; Feig et al., 2008; Soletsky and Feig, 2012).
diovascular disease, and renal disease, and the risk of morbidity and However, there is insufficient data regarding whether SUA level is an in-
death increases as blood pressure exceeds the optimal levels (Fujiyoshi dependent risk factor of incident hypertension and the subgroup that is
et al., 2012; Imano et al., 2009; Kondo et al., 2013; Okumura et al., associated with incident hypertension (Wu et al., 2016).
2014; Tozawa et al., 2003). However, approximately 90% hypertension We conducted a large-scale long-term cohort study to investigate
cases are of essential hypertension, and the etiology of its onset is unclear the relationships of SUA levels and incident hypertension and of sub-
(Anderson et al., 1994; Omura et al., 2004; Rossi et al., 2006). groups associated with incident hypertension in young and middle-
Uric acid (UA) is the final metabolite of purines in human and increas- aged Japanese males free from hypertension or diagnosed cardiovascu-
ing serum UA (SUA) levels are known to be associated with incident hy- lar disease at baseline. This study was approved by the ethics committee
pertension (Forman et al., 2009; Grayson et al., 2011; Krishnan et al., of the Nagoya University School of Medicine and all subjects gave their
informed consent for participation.

2. Methods
Abbreviations: SUA, serum uric acid; BMI, body mass index; UA, uric acid; HR, hazard
ratio; SD, standard deviation; XO, xanthine oxidase; NO, nitric oxide.
⁎ Corresponding author at: Department of Advanced Medicine in Cardiopulmonary 2.1. Study population
Disease, Nagoya University Graduate School of Medicine, Nagoya, Japan; 65 Tsurumai-
cho, Showa-ku, Nagoya, Aichi, Japan. The subjects were 33,942 Japanese male workers aged 18–60 years, recruit-
E-mail address: [email protected] (T. Kondo). ed in 2000, who underwent annual medical checkups at their workplaces from

http://dx.doi.org/10.1016/j.ypmed.2016.07.007
0091-7435/© 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
202 Y. Yokoi et al. / Preventive Medicine 90 (2016) 201–206

2000 to 2010. The law stipulates annual medical health examinations for all (BP-203IIIB; Colin Corporation, Tokyo, Japan). The blood pressure was mea-
workers in Japan, and none of them refused to participate. All subjects were sured twice on the right arm with intervals of 1 min, and the average value
employed by blue chip manufacturing companies in Aichi Prefecture, in the cen- was calculated as the baseline blood pressure. When a subject had frequent
ter of Japan. We excluded subjects with only baseline data (n = 769); with hy- premature contractions or atrial fibrillation, trained nurses confirmed the
pertension (≥140/90 mm Hg and/or taking antihypertensive drugs (n = 4223); blood pressure using a conventional mercury sphygmomanometer. Incident
who were taking drugs for hyperuricemia (n = 953), heart disease (n = 191), hypertension was defined as that when the newly detected blood pressure
renal disease (n = 122), lipid abnormalities (n = 1965), and diabetes mellitus was higher than 140/90 mm Hg and/or when antihypertensive drugs were
(n = 761). The final sample was 26,442 subjects (Fig. 1). initiated.

2.2. Examination
2.4. Statistical analysis
Annual medical checkups were performed from 2000 until 2010 or retire-
ment and included physical examination, blood pressure measurement, blood Statistical analyses were performed using the STATA software program ver-
test, dipstick urine test, and structured questionnaire. The questionnaire com- sion 11 (Stata Corp. College Station, TX, USA). Subjects were classified into
prised smoking status, alcohol intake, medical history, and medications. The tertiles on the basis of the SUA levels at the baseline. In the analyses, we used
smoking status was classified into three groups (current smoker, former smok- the lowest tertile of SUA as a reference group. Proteinuria was classified into
er, and never smoker), and alcohol intake was classified into two groups (habit- three groups: negative (−), trace (±), and ≥1+. Baseline characteristics were
ual and non-habitual). A habitual drinker was defined as a subject who drank analyzed using the trend test according to the tertile of SUA level (Table 1). Ac-
alcohol every day. Weight and height were measured while the subject was cording to the presence of incident hypertension, continuous variables were
wearing light clothing and no shoes. The body mass index was computed as compared using t-test, and categorical variables were compared using a chi-
the weight in kilograms divided by the square of the height in meters. Blood square test (Table 2).
and urine test samples were collected in the morning after fasting for 1 night. The Kaplan–Meier method was used to create the survival curves for each
Blood samples were analyzed using autoanalyzer (Dimension RxL MAX; SIE- group and compared using the log-rank test. Cox proportional hazards model
MENS, Munich, Germany) and measured by the uricase ultraviolet method. Uri- was used to calculate adjusted hazard ratio (HR) of incident hypertension ac-
nalysis was performed using a dipstick. Urinalysis for proteinuria was cording to the tertile of SUA level in five models. Model 1 was crude, model 2
conducted using Uropaper III (Eiken Chemical Co., Ltd., Tokyo, Japan), and the was further adjusted for age, model 3 was further adjusted for BMI, model 4
results were measured using a US-2100 Automated Urine Analyzer (Eiken was further adjusted for systolic blood pressure of baseline, and lastly, model
Chemical Co., Ltd., Tokyo, Japan) [trace (±), proteinuria ≥ 15 mg/dL; 1 +, 5 was further adjusted for total cholesterol, triglyceride, fasting blood glucose,
≥30 mg/dL; 2+, ≥100 mg/dL; 3+, ≥300 mg/dL; 4+, ≥1000 mg/dL]. creatinine, urine protein, smoking history, and alcohol intake.
In subgroup analysis, we stratified age (≥40 years and b 40 years) and BMI
2.3. Blood pressure measurement (≥25 kg/m2 and b 25 kg/m2) and also performed the test of interaction using
likelihood ratio test. Cut-off value of BMI was decided as per the Japanese defi-
Blood pressure was measured annually with the participant in the sit- nition of obesity (Chin and Miyazaki, 2009). All p values were two tailed, and
ting position after a 5-min rest using an automated sphygmomanometer p b 0.05 were considered statistically significant.

Fig. 1. Flow of study participant selection. Pre existing hypertension was defined as blood pressure ≥ 140/90 mm Hg and/or use of antihypertensive drugs. SUA denotes serum uric acid.
 Y. Yokoi et al. / Preventive Medicine 90 (2016) 201–206 203

Table 1
Baseline characteristics according to the tertile of SUA concentration.

Baseline SUA tertile (mg/dL)

0.1–5.3 5.4–6.2 6.3–11.6

All (Mean, 4.6) (Mean, 5.8) (Mean, 7.0) p for trend

Number of participants 26,442 8926 8935 8581

Age (years) 36.8 ± 8.7 36.9 ± 8.9 36.4 ± 8.6 37.1 ± 8.5 0.071
Body mass index (kg/m2) 22.5 ± 2.8 21.7 ± 2.6 22.3 ± 2.7 23.5 ± 2.9 b0.001
Systolic blood pressure (mm Hg) 113.9 ± 10.3 112.7 ± 10.4 113.8 ± 10.3 115.2 ± 10.1 b0.001
Diastolic blood pressure (mm Hg) 69.1 ± 9.4 68.0 ± 9.3 68.8 ± 9.2 70.6 ± 9.4 b0.001
Total cholesterol (mg/dL) 189.1 ± 31.3 184.1 ± 30.2 187.8 ± 30.9 195.7 ± 31.7 b0.001
HDL cholesterol (mg/dL) 56.9 ± 13.6 58.0 ± 13.5 57.1 ± 13.5 55.5 ± 13.7 b0.001
Triglyceride (mg/dL) 116.2 ± 75.8 103.6 ± 62.4 112.4 ± 73.3 133.2 ± 87.0 b0.001
Creatinine (mg/dL) 0.93 ± 0.11 0.91 ± 0.10 0.93 ± 0.10 0.96 ± 0.11 b0.001
Fasting blood glucose (mg/dL) 90.2 ± 12.5 89.5 ± 13.8 89.8 ± 12.0 91.3 ± 11.6 b0.001
Dipstick proteinuria
Negative 25,922 (98.0%) 8767 (98.2%) 8770 (98.2%) 8385 (97.7%)
Trace 177 (0.7%) 65 (0.7%) 55 (0.6%) 57 (0.7%)
≥1+ 343 (1.3%) 94 (1.1%) 110 (1.2%) 139 (1.6%) 0.016
Smoker
Never 6878 (26.0%) 2199 (24.6%) 2319 (26.0%) 2360 (27.5%)
Former 4938 (18.7%) 1490 (16.7%) 1631 (18.3%) 1817 (21.2%)
Current 14,626 (55.3%) 5237 (58.7%) 4985 (55.8%) 4404 (51.3%) b0.001
Alcohol intake
Habitual 9538 (36.1%) 2649 (29.7%) 3181 (35.6%) 3708 (43.2%)
Non-habitual 16,904 (63.9%) 6277 (70.3%) 5754 (64.4%) 4873 (56.8%) b0.001

HDL, high density lipoprotein; SUA, serum uric acid.

3. Result As for the baseline characteristics according to the presence of inci-

dent hypertension, mean SUA level (±SD) was 5.92 (±1.14) mg/dL in
3.1. Baseline characteristics incident hypertension and 5.67 (±1.08) mg/dL in non-hypertension.
All parameters showed statistically significant difference (Table 2).
The baseline SUA level in this population showed a Gaussian distri-
bution, and the average value [± standard deviation (SD)] was 5.78 3.2. Serum uric acid and risk for incident hypertension
(±1.11) mg/dL. The average SUA levels (mg/dL) of each group (±SD)
were 4.6 (± 0.66), 5.8 (± 0.26), and 7.0 (± 0.62) at the 1st, 2nd, and The mean follow-up period was 7.2 years, and 11,361 (43.0%) sub-
3rd tertiles, respectively. The average age and BMI (±SD) of all partici- jects developed newly detected hypertension. Newly detected hyper-
pants were 36.8 (±8.7) years and 22.5 (±2.8) kg/m2, respectively. In tension incidence was 37.4% in the 1st tertile, 41.0% in the 2nd tertile,
comparison, according to the tertile of SUA level, there were statistically and 50.8% in the 3rd tertile (log-rank test p b 0.0001; Fig. 3).
significant trends in all parameters, except for age. There was an oppo- HR [95% confidence interval (CI)] of incident hypertension of the 3rd
site trend in high-density lipoprotein cholesterol and smoking habit tertile in model 1 was 1.51 (1.45–1.58) and remained significant even
(Fig. 2, Table 1). when adjusting for age in model 2 (adjusted HR: 1.52, 95% CI: 1.45–

Table 2
Baseline characteristics according to the incident hypertension.

Incident hypertension

− + p

Number of participants (%) 15,081 (57.0%) 11,361 (43.0%)

Serum uric acid concentration (mg/dL) 5.67 ± 1.08 5.92 ± 1.14 b0.0001
Age (years) 36.4 ± 8.9 37.3 ± 8.3 b0.0001
Body mass index (kg/m2) 21.9 ± 2.5 23.3 ± 3.0 b0.0001
Systolic blood pressure (mm Hg) 110.6 ± 9.7 118.2 ± 9.5 b0.0001
Diastolic blood pressure (mm Hg) 66.7 ± 8.8 72.3 ± 9.1 b0.0001
Total cholesterol (mg/dL) 186.4 ± 31.2 192.7 ± 31.0 b0.0001
HDL cholesterol (mg/dL) 57.5 ± 13.5 56.1 ± 13.6 b0.0001
Triglyceride (mg/dL) 110.4 ± 70.7 123.9 ± 81.4 b0.0001
Creatinine (mg/dL) 0.93 ± 0.10 0.94 ± 0.11 0.0006
Fasting blood glucose (mg/dL) 89.3 ± 12.4 91.4 ± 12.6 b0.0001
Dipstick proteinuria
Negative 14,839 (98.4%) 11,083 (97.6%)
Trace 90 (0.6%) 87 (0.8%)
≥1+ 152 (1.0%) 191 (1.7%) b0.001
Smoker
Never 4090 (27.1%) 2788 (24.5%)
Former 2789 (18.5%) 2149 (18.9%)
Current 8202 (54.4%) 6424 (56.5%) b0.001
Alcohol intake
Habitual 4963 (32.9%) 4575 (40.3%)
Non-habitual 10,118 (67.1%) 6786 (59.7%) b0.001

HDL, high density lipoprotein.

204 Y. Yokoi et al. / Preventive Medicine 90 (2016) 201–206

3.3.2. BMI
As a reference, in those with BMI b 25 kg/m2 and lowest tertile of
SUA, adjusted HR in model 5 was calculated. Adjusted HR was higher
in those with BMI ≥ 25 kg/m2 than those with BMI b 25 kg/m2 but not
significant (p for interaction = 0.080; Fig. 5).

4. Discussion

In this study, we performed a large-scale long-term cohort study and

showed that SUA level is an independent predictor of incident hyper-
tension in young and middle-aged Japanese men. Furthermore, the rela-
tionship was particularly strong in those aged ≥40 years.
The mechanism by which elevated SUA levels induce hypertension
remains elusive. Oxidative stress, inflammation, nitric oxide (NO) pro-
duction impairment, vascular endothelial dysfunction, vascular smooth
muscle proliferation, and renin angiotensin system enhancement have
been reported as mechanisms for developing hypertension by hyperuri-
cemia (Corry et al., 2008; Glantzounis et al., 2005; Kang et al., 2005a,
Fig. 2. Histogram of serum uric acid concentration. SUA denotes serum uric acid and SD
2005b; Khosla et al., 2005; Kono et al., 2010).
denotes standard deviation.
Crystallization of UA itself has also been reported to cause inflamma-
tion, gouty kidney, and urinary tract calculi, and progression to renal
failure (Chonchol et al., 2007; Iseki et al., 2004; Low and Stoller, 1997;
Weiner et al., 2008). Vascular disorders are important as a mechanism
of incident hypertension by hyperuricemia, and there have been two re-
ported pathways in previous studies (Battelli et al., 2014; Kang et al.,
2005a, 2005b; Neogi et al., 2012; Price et al., 2006).
Vascular disorders can be caused by active oxygen produced by xan-
thine oxidase (XO). Hypoxanthine is a UA precursor and is metabolized
to UA by XO present in the vascular endothelium where the active oxy-
gen generated in the process inactivates NO in the vascular endothelium
and proliferates vascular smooth muscle thus promoting arteriosclero-
sis (Battelli et al., 2014; Neogi et al., 2012).
Another pathway causing vascular disorders is the mechanism by
UA itself flowing into the cells via a UA transporter present in the vascu-
lar endothelial cells and vascular smooth muscle cells. The UA transport-
er is mainly present in the proximal tubule in addition to the vascular
Fig. 3. Cumulative incidence of hypertension according to the tertile of the serum uric acid smooth muscle cells (Price et al., 2006). UA causes vascular smooth
concentration estimated using the Kaplan–Meier method. Mean follow-up time was muscle cell proliferation through urate transporters (Kang et al.,
7.2 years. The number of cumulative incident hypertension episodes was 11,361 2005a, 2005b).
(43.0%). The blue dotted line indicates the 1st tertile group, green dash indicates the
2nd tertile group, and red solid indicates the 3rd tertile group.
The vasorelaxant effect of acetylcholine was reported to attenuate as
UA concentration increased in in vitro experiments using rat aorta
(Nakagawa et al., 2006).
In developing hypertension by hyperuricemia, two phases would be
conceived. In the first phase, inhibition of NO production and activation
1.59) and remained statistically significant in model 5, although HRs
of the renin–angiotensin system by hyperuricemia causes excessive va-
were attenuated as further adjusting for other covariates (adjusted
soconstriction, resulting in hypertension. This process is reversible and
HR: 1.15, 95% CI: 1.10–1.21; Table 3).
is a UA-dependent reaction where blood pressure can be reduced by
lowering UA (Feig, 2012; Mazzali et al., 2001; Mazzali et al., 2002).
3.3. Subgroup analysis In the second phase, when hyperuricemia persists, the proliferation
of vascular smooth muscle cells thickens vessel wall and increases blood
3.3.1. Age pressure because of changes in the vascular structure. This step is UA in-
Adjusted HR in model 5 was calculated as a reference for those aged dependent and blood pressure does not return to the original level any-
b40 years and the lowest tertile of SUA levels. Adjusted HR was signifi- more by lowering UA as in the first phase (Feig, 2012; Kang et al., 2005a,
cantly higher in those aged ≥40 years than those aged b40 years (p for 2005b). In this study, a stronger relationship between SUA level and hy-
interaction = 0.035; Fig. 4). pertension developed in subjects aged ≥40 years. Long-term exposure

Table 3
Hazard ratios for incident hypertension.

Hazard ratio (95% CI)

SUA tertile (mg/dL) Model 1 Model 2 Model 3 Model 4 Model 5

1st 0.1–5.3 1 (reference) 1 (reference) 1 (reference) 1 (reference) 1 (reference)

2nd 5.4–6.2 1.12 (1.06–1.17) 1.13 (1.08–1.18) 1.04 (1.00–1.09) 1.01 (0.96–1.06) 1.01 (0.96–1.05)
3rd 6.3–11.6 1.51 (1.45–1.58) 1.52 (1.45–1.59) 1.22 (1.16–1.27) 1.16 (1.11–1.22) 1.15 (1.10–1.21)

Model 1, crude; Model 2, age; Model 3: Model 2 + BMI; Model 4: Model 3 + baseline BP; Model 5: Model 4 + total cholesterol, triglyceride, creatinine, proteinuria, fasting blood sugar,
smoking, alcohol intake. CI, confidence interval; SUA, serum uric acid.
 Y. Yokoi et al. / Preventive Medicine 90 (2016) 201–206 205

Fig. 4. Adjusted hazard ratios stratified by age (b40 years vs. ≥40 years). Adjusted hazard ratios were calculated in Model 5. The lowest tertile of serum uric acid and age ≤ 40 years group
was used as reference. Covariates included age, body mass index, baseline blood pressure, total cholesterol, triglyceride, creatinine, proteinuria, fasting blood sugar, smoking, and alcohol
intake in Model 5.

to hyperuricemia may progress to irreversible arteriosclerosis of blood heart disease, renal disease, lipid abnormalities, and diabetes mellitus.
vessels, which could be related to higher HR in ≥ 40-year-old subjects This association was stronger among subjects ≥40 years old.
and existence of interaction between age and SUA.
On the other hand, there was no significant interaction between SUA Conflicts of interest
and BMI in this study. SUA and obesity may resonate each other by ac-
tivating the renin-angiotensin system (Corry et al., 2008; Goossens et We declare no conflicts of interests.
al., 2003; Reaven, 2011). In fact, the Tromsø Study in Norway, in con-
trast to our present study, demonstrated that higher baseline SUA was Transparency document
associated with higher odds of developing elevated blood pressure in
overweight subjects (BMI ≥ 25 kg/m2), but not in normal-weight sub- The Transparency document associated to this article can be found,
jects (BMI b 25 kg/m2) (Norvik et al., 2016). The subjects in the present in online version.
study were relatively thin and the average of BMI was 22.5 kg/m2.
Therefore, the significant interaction between SUA and BMI (b25 kg/ Acknowledgments
m2 vs. ≥ 25 kg/m2) would not be observed. Further research of more
overweight/obese samples is needed to verify the interaction. We thank the healthcare providers for their hard work and excellent
assistance in this study.
5. Limitations
References
There were some limitations to this study. Only Japanese men were Agarwal, V., Hans, N., Messerli, F.H., 2013. Effect of allopurinol on blood pressure: a sys-
used as subjects because there was not long-term enough data for fe- tematic review and meta-analysis. J. Clin. Hypertens. (Greenwich) 15, 435–442.
male workers; the number of female workers was limited and most fe- Anderson Jr., G.H., Blakeman, N., Streeten, D.H., 1994. The effect of age on prevalence of
secondary forms of hypertension in 4429 consecutively referred patients.
male workers took one to three years maternity leave, during which J. Hypertens. 12, 609–615.
blood pressures were not followed. Blood pressure was measured only Battelli, M.G., Polito, L., Bolognesi, A., 2014. Xanthine oxidoreductase in atherosclerosis
once a year; in addition, blood pressure was not monitored at home. pathogenesis: not only oxidative stress. Atherosclerosis 237, 562–567.
Chin, R., Miyazaki, S., 2009. Criteria of obesity and obesity disease in Japan. Nihon Rinsho
Data on abdominal circumference, daily exercise, diet, stress, and family
67, 297–300.
history were not collected. Chonchol, M., Shlipak, M.G., Katz, R., Sarnak, M.J., Newman, A.B., Siscovick, D.S., et al.,
2007. Relationship of uric acid with progression of kidney disease. Am. J. Kidney
Dis. 50, 239–247.
6. Conclusion Corry, D.B., Eslami, P., Yamamoto, K., Nyby, M.D., Makino, H., Tuck, M.L., 2008. Uric acid
stimulates vascular smooth muscle cell proliferation and oxidative stress via the vas-
cular renin-angiotensin system. J. Hypertens. 26, 269–275.
High SUA levels increases the risk of incident hypertension in young Feig, D.I., 2012. The role of uric acid in the pathogenesis of hypertension in the young.
and middle-aged Japanese males without hypertension, hyperuricemia, J. Clin. Hypertens. (Greenwich) 14, 346–352.

Fig. 5. Adjusted hazard ratios stratified by body mass index (BMI; b25 kg/m2 vs. ≥25 kg/m2). Adjusted hazard ratios were calculated in Model 5. The lowest tertile of serum uric acid and
BMI b 25 kg/m2 group was used as reference. Covariates included age, body mass index, baseline blood pressure, total cholesterol, triglyceride, creatinine, proteinuria, fasting blood sugar,
smoking, and alcohol intake in Model 5.
206 Y. Yokoi et al. / Preventive Medicine 90 (2016) 201–206

Feig, D.I., Soletsky, B., Johnson, R.J., 2008. Effect of allopurinol on blood pressure of adoles- Mellen, P.B., Bleyer, A.J., Erlinger, T.P., Evans, G.W., Nieto, F.J., Wagenknecht, L.E., et al.,
cents with newly diagnosed essential hypertension: a randomized trial. JAMA 300, 2006. Serum uric acid predicts incident hypertension in a biethnic cohort: the athero-
924–932. sclerosis risk in communities study. Hypertension 48, 1037–1042.
Forman, J.P., Choi, H., Curhan, G.C., 2009. Uric acid and insulin sensitivity and risk of inci- Nakagawa, T., Hu, H., Zharikov, S., Tuttle, K.R., Short, R.A., Glushakova, O., et al., 2006. A
dent hypertension. Arch. Intern. Med. 169, 155–162. causal role for uric acid in fructose-induced metabolic syndrome. Am. J. Phys. Renal
Fujiyoshi, A., Ohkubo, T., Miura, K., Murakami, Y., Nagasawa, S.Y., Okamura, T., et al., 2012. Phys. 290, F625–F631.
Blood pressure categories and long-term risk of cardiovascular disease according to Nakanishi, N., Okamoto, M., Yoshida, H., Matsuo, Y., Suzuki, K., Tatara, K., 2003. Serum uric
age group in Japanese men and women. Hypertens. Res. 35, 947–953. acid and risk for development of hypertension and impaired fasting glucose or Type II
Glantzounis, G.K., Tsimoyiannis, E.C., Kappas, A.M., Galaris, D.A., 2005. Uric acid and oxida- diabetes in Japanese male office workers. Eur. J. Epidemiol. 18, 523–530.
tive stress. Curr. Pharm. Des. 11, 4145–4151. Neogi, T., George, J., Rekhraj, S., Struthers, A.D., Choi, H., Terkeltaub, R.A., 2012. Are either
Goossens, G.H., Blaak, E.E., van Baak, M.A., 2003. Possible involvement of the adipose tis- or both hyperuricemia and xanthine oxidase directly toxic to the vasculature? A crit-
sue renin-angiotensin system in the pathophysiology of obesity and obesity-related ical appraisal. Arthritis Rheum. 64, 327–338.
disorders. Obes. Rev. 4, 43–55. Norvik, J.V., Storhaug, H.M., Ytrehus, K., Jenssen, T.G., Zykova, S.N., Eriksen, B.O., Solbu,
Grayson, P.C., Kim, S.Y., LaValley, M., Choi, H.K., 2011. Hyperuricemia and incident hyper- M.D., 2016. Overweight modifies the longitudinal association between uric acid and
tension: a systematic review and meta-analysis. Arthritis Care Res. 63, 102–110. some components of the metabolic syndrome: the Tromsø Study. BMC Cardiovasc.
Imano, H., Kitamura, A., Sato, S., Kiyama, M., Ohira, T., Yamagishi, K., et al., 2009. Trends for Disord. 16, 85.
blood pressure and its contribution to stroke incidence in the middle-aged Japanese Okumura, N., Kondo, T., Matsushita, K., Osugi, S., Shimokata, K., Matsudaira, K., et al., 2014.
population: the Circulatory Risk in Communities Study (CIRCS). Stroke 40, Associations of proteinuria and the estimated glomerular filtration rate with incident
1571–1577. hypertension in young to middle-aged Japanese males. Prev. Med. 60, 48–54.
Iseki, K., Ikemiya, Y., Inoue, T., Iseki, C., Kinjo, K., Takishita, S., 2004. Significance of hyper- Omura, M., Saito, J., Yamaguchi, K., Kakuta, Y., Nishikawa, T., 2004. Prospective study on
uricemia as a risk factor for developing ESRD in a screened cohort. Am. J. Kidney Dis. the prevalence of secondary hypertension among hypertensive patients visiting a
44, 642–650. general outpatient clinic in Japan. Hypertens. Res. 27, 193–202.
Kang, D.H., Han, L., Ouyang, X., Kahn, A.M., Kanellis, J., Li, P., et al., 2005a. Uric acid causes Price, K.L., Sautin, Y.Y., Long, D.A., Zhang, L., Miyazaki, H., Mu, W., et al., 2006. Human vas-
vascular smooth muscle cell proliferation by entering cells via a functional urate cular smooth muscle cells express a urate transporter. J. Am. Soc. Nephrol. 17,
transporter. Am. J. Nephrol. 25, 425–433. 1791–1795.
Kang, D.H., Park, S.K., Lee, I.K., Johnson, R.J., 2005b. Uric acid-induced C-reactive protein Reaven, G.M., 2011. Insulin resistance: the link between obesity and cardiovascular dis-
expression: implication on cell proliferation and nitric oxide production of human ease. Med. Clin. North Am. 95, 875–892.
vascular cells. J. Am. Soc. Nephrol. 16, 3553–3562. Rossi, G.P., Bernini, G., Caliumi, C., Desideri, G., Fabris, B., Ferri, C., et al., 2006. A prospective
Khosla, U.M., Zharikov, S., Finch, J.L., Nakagawa, T., Roncal, C., Mu, W., et al., 2005. Hyper- study of the prevalence of primary aldosteronism in 1,125 hypertensive patients.
uricemia induces endothelial dysfunction. Kidney Int. 67, 1739–1742. J. Am. Coll. Cardiol. 48, 2293–2300.
Kondo, T., Osugi, S., Shimokata, K., Honjo, H., Okumura, N., Matsudaira, K., et al., 2013. Car- Soletsky, B., Feig, D.I., 2012. Uric acid reduction rectifies prehypertension in obese adoles-
diovascular events increased at normal and high-normal blood pressure in young and cents. Hypertension 60, 1148–1156.
middle-aged Japanese male smokers but not in nonsmokers. J. Hypertens. 31, Taniguchi, Y., Hayashi, T., Tsumura, K., Endo, G., Fujii, S., Okada, K., 2001. Serum uric acid
263–270. and the risk for hypertension and Type 2 diabetes in Japanese men: the Osaka Health
Kono, H., Chen, C.J., Ontiveros, F., Rock, K.L., 2010. Uric acid promotes an acute inflamma- Survey. J. Hypertens. 19, 1209–1215.
tory response to sterile cell death in mice. J. Clin. Invest. 120, 1939–1949. Tozawa, M., Iseki, K., Iseki, C., Kinjo, K., Ikemiya, Y., Takishita, S., 2003. Blood pressure pre-
Krishnan, E., Kwoh, C.K., Schumacher, H.R., Kuller, L., 2007. Hyperuricemia and incidence dicts risk of developing end-stage renal disease in men and women. Hypertension 41,
of hypertension among men without metabolic syndrome. Hypertension 49, 1341–1345.
298–303. Weiner, D.E., Tighiouart, H., Elsayed, E.F., Griffith, J.L., Salem, D.N., Levey, A.S., 2008. Uric
Low, R.K., Stoller, M.L., 1997. Uric acid-related nephrolithiasis. Urol. Clin. North Am. 24, acid and incident kidney disease in the community. J. Am. Soc. Nephrol. 19,
135–148. 1204–1211.
Mazzali, M., Hughes, J., Kim, Y.G., Jefferson, J.A., Kang, D.H., Gordon, K.L., et al., 2001. Ele- Wu, A.H., Gladden, J.D., Ahmed, M., Ahmed, A., Filippatos, G., 2016. Relation of serum uric
vated uric acid increases blood pressure in the rat by a novel crystal-independent acid to cardiovascular disease. Int. J. Cardiol. 213, 4–7.
mechanism. Hypertension 38, 1101–1106.
Mazzali, M., Kanellis, J., Han, L., Feng, L., Xia, Y.Y., Chen, Q., et al., 2002. Hyperuricemia in-
duces a primary renal arteriolopathy in rats by a blood pressure-independent mech-
anism. Am. J. Phys. Renal Phys. 282, F991–F997.