Professional Documents
Culture Documents
000499867
000499867
78 Saemunan-gil, Jongro-Gu
E-Mail karger@karger.com
Seoul 110-746 (Republic of Korea)
www.karger.com/crd
Downloaded by:
2 Cardiology Jung/Oh/Choi/Ryoo/Chung/Hong/Park
DOI: 10.1159/000499867
Lund University Libraries
Downloaded by:
Table 1. Main clinical characteristics of the study population according to SBP
SBP (mm Hg): <110 110–119 120–129 130–139 140–149 ≥150 p for
(n = 2,072) (n = 1,633) (n = 1,060) (n = 501) (n = 283) (n = 260) trend
Male sex 864 (41.7) 852 (52.2) 622 (58.7) 299 (59.7) 158 (55.8) 137 (52.7) <0.001
Age, years 48.1±7.5 50.0±8.2 52.5±8.8 54.6±9.3 56.6±8.7 57.9±8.1 <0.001
Fasting glucose, mg/dL 82.7±14.6 83.9±14.3 85.3±16.3 84.4±13.8 84.3±10.1 84.8±11.7 <0.001
Total cholesterol, mg/dL 184.7±32.8 188.6±34.0 190.9±35.4 187.7±37.2 193.0±39.9 192.1±35.7 <0.001
HDL cholesterol, mg/dL 45.8±9.7 46.5±10.2 47.4±9.7 49.3±10.1 50.9±10.4 51.5±11.6 <0.001
TG, mg/dL 128.0±78.8 138.0±84.6 139.2±87.1 123.4±49.0 128.8±53.7 127.7±60.4 0.923
Average alcohol intake, g/day 6.6±16.2 10.4±24.6 12.8±24.6 11.6±24.0 9.8±18.0 10.0±18.7 <0.001
Current smokers, % 26.1 27.5 28.1 27.1 26.5 24.6 0.903
Regular exercise, % 31.8 40.7 43.8 48.3 49.8 51.9 <0.001
BMI 23.5±2.7 24.0±2.8 23.9±2.9 23.4±2.9 23.6±2.6 23.8±3.2 0.247
WC, cm 78.1±7.7 80.2±7.4 81.2±7.5 79.9±7.0 80.2±7.1 80.7±7.4 <0.001
SBP, mm Hg 101.0±6.3 114.2±2.9 123.9±2.9 133.6±2.9 143.8±3.0 161.2±11.5 <0.001
DBP, mm Hg 68.8±6.6 76.7±5.6 81.6±6.5 87.8±7.3 92.2±8.4 98.1±9.4 <0.001
History of hypertension, % 0.8 2.7 7.0 17.6 29.0 48.5 <0.001
Diabetes mellitus, % 3.7 6.2 5.8 4.4 5.3 6.5 0.039
Incidence of MetS 387 (18.7) 511 (31.3) 408 (38.5) 182 (36.3) 129 (45.6) 112 (43.1) <0.001
Continuous variables are expressed as means ± SD, and categorical variables are expressed as numbers (%) unless otherwise stated.
stolic BP (DBP). The WC, height, and weight were also Statistical Analysis
measured and the BMI was calculated for all of the par- The study participants were categorized according to
ticipants. SBP and DBP, respectively. Data are presented as means
After fasting overnight for 12 h, the plasma concentra- ± SD within study groups for continuous variables and as
tions of glucose, total cholesterol, TG, and HDL choles- proportions for categorical variables. The linear trends of
terol were measured enzymatically using a Hitachi Auto- variables between study groups were calculated using a
matic Analyzer 7600 (Hitachi, Tokyo, Japan). Fasting linear regression model for continuous variables and the
plasma insulin concentrations were determined using a Cochran-Armitage trend test for categorical variables.
radioimmunoassay kit (Linco Research, St. Charles, MO, The baseline characteristics of the study population, in-
USA). HbA1C levels were measured via high-perfor- cluding participants with MetS, are also presented in the
mance liquid chromatography (VARIANT II; Bio-Rad online supplementary Tables 1 and 2 (for all online suppl.
Laboratories, Hercules, CA, USA). material, see www.karger.com/doi/10.1159/000499867).
The presence of MetS was determined according to Crude and multivariate adjusted HR and their 95% CI
the joint interim statement of the International Diabetes for MetS were obtained using the Cox proportional haz-
Federation (IDF) Task Force on Epidemiology and Pre- ards model. The covariates of the adjusted model were:
vention [1]. Elevated BP was defined as SBP or DBP age, sex, area, regular exercise, WC, diabetes, smoking,
≥130/85 mm Hg, elevated fasting serum glucose was de- alcohol intake, and HDL cholesterol. The incidence cases
fined as levels ≥100 mg/dL, high serum TG was defined of MetS and the incidence density (incidence cases per
as levels ≥150 mg/dL, low HDL cholesterol was defined 1,000 person-years) were also calculated in each SBP and
as levels <40 mg/dL in men and <50 mg/dL in women. DBP group. In the analysis of all of the participants, SBP
The presence of visceral obesity was defined based on the and DBP were divided by a unit of 10 mm Hg, and the
criteria of the Korean Society for the Study of Obesity group with the lowest BP level was used as the reference
(i.e., WC ≥90 cm for men and ≥85 cm for women) [17]. group (i.e., SBP: <110, 110–119, 120–129, 130–139, 140–
MetS was defined as the presence of 3 or more of the 149, and ≥150 mm Hg; DBP: <70, 70–79, 80–89, 90–99,
above components. and ≥100 mm Hg). The subgroup analyses were per-
formed in normotensive participants (subjects with a BP
130.235.66.10 - 7/21/2019 7:24:35 PM
DBP (mm Hg): <70 70–79 80–89 90–100 ≥100 p for trend
(n = 1,191) (n = 2,292) (n = 1,548) (n = 579) (n = 199)
Male sex 388 (32.6) 1,147 (50.0) 898 (58.0) 376 (64.9) 123 (61.8) <0.001
Age, years 48.3±7.7 50.2±8.4 52.2±8.9 54.2±9.0 53.4±8.6 <0.001
Fasting glucose, mg/dL 81.8±14.5 84.1±15.6 84.6±13.9 84.4±12.2 85.5±10.2 <0.001
Total cholesterol, mg/dL 183.3±32.1 186.5±33.9 190.8±35.1 191.5±37.2 200.1±39.8 <0.001
HDL cholesterol, mg/dL 46.3±9.8 46.1±9.9 47.6±10.3 49.7±9.7 51.3±11.6 <0.001
TG, mg/dL 124.5±71.8 133.0±79.2 137.8±81.1 132.8±84.1 132.4±65.8 0.003
Average alcohol intake, g/day 5.4±15.8 8.7±20.7 11.4±22.5 15.6±29.1 12.0±18.9 <0.001
Current smokers, % 20.7 27.6 29.7 30.6 24.1 <0.001
Regular exercise, % 29.0 40.2 43.4 48.4 43.7 <0.001
BMI 23.2±2.6 23.8±2.8 23.9±2.9 24.0±2.7 24.1±2.9 <0.001
WC, cm 76.9±7.5 79.7±7.5 80.7±7.5 81.5±6.9 81.1±7.0 <0.001
SBP, mm Hg 100.5±9.2 111.7±8.8 122.9±10.7 138.2±12.5 155.0±15.9 <0.001
DBP, mm Hg 63.9±4.4 74.2±3.1 83.3±2.9 93.2±2.9 104.8±5.7 <0.001
History of hypertension, % 0.8 2.6 8.7 24.4 43.7 <0.001
Diabetes mellitus, % 3.9 5.9 5.0 4.8 4.0 0.810
Incidence of MetS 192 (16.1) 636 (27.7) 575 (37.1) 237 (40.9) 89 (44.7) <0.001
Continuous variables are expressed as mean ± SD, and categorical variables are expressed as numbers (%) unless otherwise stated.
<140/90 mm Hg and without a history of hypertension). observed in the DBP groups (Table 2). The higher-DBP
The normotensive participants were grouped by a unit of groups also had the more males and worse metabolic pro-
10 mm Hg in SBP and a unit of 5 mm Hg in DBP (i.e., files than the lowest-DBP group, besides HDL cholesterol
SBP: <110, 110–119, and 120–129 mm Hg; DBP: <70, 70– and regular exercise. Statistical significance was identified
74, and 75–79 mm Hg). All statistical analyses were per- for all clinical and biochemical characteristics.
formed using R 3.4.3 (R Foundation for Statistical Com- Table 3 shows the unadjusted and adjusted HR and
puting, Vienna, Austria), p < 0.05 was considered statisti- 95% CI for MetS in the SBP and DBP groups. Even after
cally significant in all analyses. adjusting for the metabolic covariates, elevated SBP and
DBP levels were significantly associated with increased
HR for MetS. A clear dose-response relationship was ob-
Results served between DBP levels and HR for MetS (< 70 mm
Hg: reference, 70–79 mm Hg: HR = 1.46 [95% CI 1.24–
A total of 5,809 participants (2,932 men and 2,877 1.72], 80–89 mm Hg: HR = 2.10 [95% CI 1.77–2.49], 90–
women, mean age 50.8 years) were enrolled into this 99 mm Hg: HR = 2.63 [95% CI 2.16–3.21], and ≥100 mm
study. The overall incidence rate of MetS was 28.6% (n = Hg: HR = 3.94 [95% CI 3.05–5.10]). The subgroup analy-
1,729), and incidence density was 40.8. The final follow- sis for normotensive participants also indicated increased
up examination was done in 2011–2012. HR for MetS proportionally to SBP (<110 mm Hg: refer-
Table 1 presents the baseline clinical and biochemical ence, 110–119 mm Hg: HR = 1.60 [95% CI 1.40–1.84],
characteristics of the study participants according to SBP. and 120–129 mm Hg: HR = 2.12 [95% CI 1.82–2.48]) and
Compared to the group with the lowest SBP level, the high- DBP levels (<70 mm Hg: reference, 71–74 mm Hg: HR =
er-SBP groups tended to have worse metabolic profiles in- 1.31 [95% CI 1.09–1.58], and 75–79 mm Hg: HR = 1.51
cluding age, fasting glucose, WC, DBP, prevalence of DM, [95% CI 1.25–1.81]) (Table 4).
and incidence of MetS. The rates of male gender, HDL cho- Online supplementary Tables 1 and 2 present the as-
lesterol, and regular exercise were higher in the higher-SBP sociation between BP levels and the risk of incident MetS
groups than in the lowest-SBP group. However, statistical based on the cut-off of WC from the IDF task force (WC
significance was not identified for TG, BMI, and the pro- ≥102 cm in men and ≥89 cm in women). Even after
portion of current smokers. These findings were similarly adopting a cut-off for WC that was higher than the mod-
130.235.66.10 - 7/21/2019 7:24:35 PM
4 Cardiology Jung/Oh/Choi/Ryoo/Chung/Hong/Park
DOI: 10.1159/000499867
Lund University Libraries
Downloaded by:
Table 3. HR and 95% CI for the incidence of MetS
Values were adjusted for age, sex, area, regular exercise, WC, smoking, HDL cholesterol, diabetes mellitus,
and alcohol intake. a Incidence cases per 1,000 person-years.
Table 4. HR and 95% CI for the incidence of MetS in subjects without hypertension
Adjusted for age, sex, area, regular exercise, WC, smoking, HDL cholesterol, diabetes mellitus, and alcohol
intake. a Incidence cases per 1,000 person-years.
ified cut-off in Koreans, the risk of MetS significantly in- yielded a mean BMI of less than 25 and a mean WC of 85
creased proportionally to the levels of baseline SBP and cm. Additionally, none of the groups had a mean BMI
DBP in both all and the normotensive participants. ≥25 or a WC ≥85 cm. These findings show the need to
assess other predictors of MetS besides obesity in Asians.
We grouped the study participants by stratification of
Discussion SBP and DBP and evaluated the risk of incident MetS ac-
cording to BP strata. Our results, despite the generally
Obesity has a great influence on the pathogenesis of nonobese figures of the subjects, showed a proportional
both MetS and hypertension. However, it is evident that relationship between the risk of MetS and SBP and DBP.
Asians have a higher metabolic risk despite a lower degree In the SBP groups, compared to the group with a SBP
of obesity compared to Caucasians [18]. Our study also <110 mm Hg, the risk of MetS significantly increased up
130.235.66.10 - 7/21/2019 7:24:35 PM
6 Cardiology Jung/Oh/Choi/Ryoo/Chung/Hong/Park
DOI: 10.1159/000499867
Lund University Libraries
Downloaded by:
Second, we could not identify detailed information on Acknowledgement
antihypertensive medication during the follow-up. Ran- Data in this study were from the KoGES (4851–302), National
domized clinical trials have shown that antihypertensive Research Institute of Health, Centers for Disease Control and Pre-
therapies with thiazide diuretics and β-blockers are asso- vention, Ministry for Health and Welfare, Korea.
ciated with an increased incidence of new onset diabetes
and other metabolic abnormalities [28, 29]. Thus, it is in-
ferred that types of antihypertensive medications might Statement of Ethics
have an impact on incident MetS during follow-up. How- Ethical approval for the study protocol and analysis of the data
ever, it was hard to confirm the types and kinds of anti- was obtained from the institutional review board of the Kangbuk
hypertensive medications during follow-up. This limita- Samsung Hospital.
tion arose from our dependency on a self-report ques-
tionnaire for identification of a history of antihypertensive Author Contributions
medications. Thus, the absence of detailed information
on antihypertensive medications should be recognized as Ju Young Jung coordinated this study, analyzed the data, and
a limitation. wrote this paper as first author. Chang-Mo Oh played a role in
analysis of the data and verification of the results. Joong-Myung
Third, our study participants were comprised of Ko- Choi participated in English editing and review of this paper. Jae-
reans only. Since the cardiometabolic risk can vary Hong Ryoo and Pil-Wook Chung participated in the conduction
among ethnicities and regions at given metabolic condi- of this study and the writing of this paper. Hyun Pyo Hong con-
tions, it is unlikely that our findings can be extrapolated ducted additional statistical analysis and writing of this paper in
to other ethnic and regional groups. the process of revision of this work.
Sung Keun Park is the guarantor of this work and, as such, had
In conclusion, our study presented that an elevated BP full access to all of the data in this study and takes responsibility
was significantly associated with an increased risk of in- for the integrity of the data and the accuracy of the data analysis.
cident MetS in a Korean population. The incidence and All of the authors had access to the data used in this study and
HR of MetS increased proportionally to the BP level, and participated in the writing of this paper.
this was identically observed even in the normotensive
population. These findings indicate the clinical signifi-
cance of BP as a metabolic risk factor.
References
1 Alberti KG, Eckel RH, Grundy SM, Zimmet and WHR. Atherosclerosis risk in communi- 8 Cuspidi C, Meani S, Fusi V, Severgnini B,
PZ, Cleeman JI, Donato KA, et al.; Interna- ties. Ann Epidemiol. 1997 Aug;7(6):407–16. Valerio C, Catini E, et al. Metabolic syndrome
tional Diabetes Federation Task Force on Ep- 4 Laaksonen DE, Lakka HM, Salonen JT, Nis- and target organ damage in untreated essen-
idemiology and Prevention; Hational Heart, kanen LK, Rauramaa R, Lakka TA. Low levels tial hypertensives. J Hypertens. 2004 Oct;
Lung, and Blood Institute; American Heart of leisure-time physical activity and cardiore- 22(10):1991–8.
Association; World Heart Federation; Inter- spiratory fitness predict development of the 9 Schillaci G, Pirro M, Vaudo G, Gemelli F,
national Atherosclerosis Society; Internation- metabolic syndrome. Diabetes Care. 2002 Marchesi S, Porcellati C, et al. Prognostic val-
al Association for the Study of Obesity. Har- Sep;25(9):1612–8. ue of the metabolic syndrome in essential hy-
monizing the metabolic syndrome: a joint in- 5 Mokdad AH, Bowman BA, Ford ES, Vinicor pertension. J Am Coll Cardiol. 2004 May;
terim statement of the International Diabetes F, Marks JS, Koplan JP. The continuing epi- 43(10):1817–22.
Federation Task Force on Epidemiology and demics of obesity and diabetes in the United 10 Ferrannini E, Natali A, Capaldo B, Lehtovirta
Prevention; National Heart, Lung, and Blood States. JAMA. 2001 Sep;286(10):1195–200. M, Jacob S, Yki-Järvinen H; European Group
Institute; American Heart Association; World 6 O’Neill S, O’Driscoll L. Metabolic syndrome: for the Study of Insulin Resistance (EGIR). In-
Heart Federation; International Atheroscle- a closer look at the growing epidemic and its sulin resistance, hyperinsulinemia, and blood
rosis Society; and International Association associated pathologies. Obes Rev. 2015 Jan; pressure: role of age and obesity. Hyperten-
for the Study of Obesity. Circulation. 2009 16(1):1–12. sion. 1997 Nov;30(5):1144–9.
Oct;120(16):1640–5. 7 Lim S, Shin H, Song JH, Kwak SH, Kang SM, 11 Halperin RO, Sesso HD, Ma J, Buring JE,
2 Han TS, Williams K, Sattar N, Hunt KJ, Lean Won Yoon J, et al. Increasing prevalence of Stampfer MJ, Gaziano JM. Dyslipidemia and
ME, Haffner SM. Analysis of obesity and hy- metabolic syndrome in Korea: the Korean the risk of incident hypertension in men. Hy-
perinsulinemia in the development of meta- National Health and Nutrition Examination pertension. 2006 Jan;47(1):45–50.
bolic syndrome: San Antonio Heart Study. Survey for 1998-2007. Diabetes Care. 2011 12 Kaplan MN. Primary hypertension: Patho-
Obes Res. 2002 Sep;10(9):923–31. Jun;34(6):1323–8. genesis. Kaplan’s Clinical Hypertension. 8th
3 Liese AD, Mayer-Davis EJ, Tyroler HA, Davis ed. Philadelphia: Lippincott Williams &
CE, Keil U, Duncan BB, et al. Development of Wilkins; 2002. p. 109–10.
the multiple metabolic syndrome in the ARIC
cohort: joint contribution of insulin, BMI,
130.235.66.10 - 7/21/2019 7:24:35 PM
8 Cardiology Jung/Oh/Choi/Ryoo/Chung/Hong/Park
DOI: 10.1159/000499867
Lund University Libraries
Downloaded by: