Published papers

To evaluate the cardiovascular mortality risk in association with blood pressure level among people with and without antihypertensive treatment, we performed the participant-level meta-analysis that included 39 705 Japanese from 6 cohorts (58.4% women; mean age, 60.1 years; 20.4% treated). Multivariable-adjusted Cox models were used to analyze the risk of cardiovascular mortality and its subtypes among 6 blood pressure levels according to recent guidelines, optimal to Grade 3 hypertension, and the usage of antihypertensive medication at baseline. During median 10.0 years of follow-up, there were 2032 cardiovascular deaths (5.1 per 1000 person-years), of which 410 deaths were coronary heart disease, 371 were heart failure, and 903 deaths were stroke. Treated participants had significantly higher risk for cardiovascular mortality (hazard ratios, 1.50; 95% confidence intervals, 1.36-1.66), coronary heart disease (hazard ratios, 1.53; confidence intervals, 1.23-1.90), heart failure (hazard ratios, 1.39; confidence intervals, 1.09-1.76), and stroke (hazard ratios, 1.48; confidence intervals, 1.28-1.72) compared with untreated people. Among untreated participants, the risks increased linearly with an increment of blood pressure category (P≤0.011). The risk increments per blood pressure category were higher in young participants (<60 years; 22% to 79%) than those in old people (≥60 years; 7% to 15%) with significant interaction for total cardiovascular, heart failure, and stroke mortality (P≤0.026). Among treated participants, the significant linear association was also observed for cardiovascular mortality (P=0.0003), whereas no stepwise increase in stroke death was observed (P=0.19). The risks of cardiovascular mortality were ≈1.5-fold high in participants under antihypertensive medication. More attention should be paid to the residual cardiovascular risks in treated patients. © 2014 American Heart Association, Inc.

Blood pressure (BP) categories defined by systolic BP (SBP) and diastolic BP (DBP) are commonly used. However, the BP category-specific risk of cardiovascular disease (CVD) has not been thoroughly investigated in different age groups. The aim of this study was to assess long-term CVD risk and its impact according to BP categories and age group. Pooling individual data from 10 cohorts, we studied 67 309 Japanese individuals (40-89 years old) who were free of CVD at baseline: we categorized them as belonging to three age groups: middle-aged (40-64 years), elderly (65-74 years) and very elderly (75-89 years). BP was classified according to the 2009 Japanese Society of Hypertension Guidelines. Cox models were used to estimate adjusted hazard ratios for CVD deaths. We observed 1944 CVD deaths over a mean follow-up of 10.2 years. In all age groups, the overall relationship between BP category and CVD risk was positive, with a greater strength observed for younger age groups. We observed a trend of increased risk from SBP/DBP130/85 mm Hg in the very elderly, and a significant increase from SBP/DBP120/80 mm Hg in the other age groups. The population attributable fractions (PAFs) of CVD death in reference to the SBP/DBP120/80 mm Hg category ranged from 23.4% in the very elderly to 60.3% in the middle-aged. We found an overall graded increase in CVD risk with higher BP category in the very elderly. The PAFs suggest that keeping BP levels low is an important strategy for primary CVD prevention, even in an elderly population. © 2012 The Japanese Society of Hypertension All rights reserved.

Adiponectin is most abundantly expressed in adipose tissue and well known to play an important role in metabolic regulation. Several studies have attempted to identify the genetic determinants of metabolic syndrome (MetS), though no study has revealed a cis- or transsingle nucleotide polymorphism (SNP) that affects plasma adiponectin levels, except the adiponectin structure gene and genes encoding adiponectin-regulatory proteins. We performed a genome-wide association study in regards to plasma adiponectin concentrations in 3,310 Japanese subjects. We identified the strongest statistically associated SNP (rs4783244) with adiponectin levels (P = 3.8 × 10 -19) in the first intron of CDH13 (T-cadherin) gene in a 30-kb haplotype block covering the promoter region to first intron. In addition, rs12051272 SNP genotypes in linkage disequilibrium with rs4783244 were found to be more significantly associated with adiponectin levels (P = 9.5×10 -20) and specifically with the levels of high-molecular weight (HMW) adiponectin, a subtype form associated with parameters related to glucose metabolism. Our results did show more significant association with adiponectin levels than rs12444338 (in CDH13) SNP genotypes reported recently. We suggest that the phenotype-affecting haplotype tagged by rs12051272 SNP would affect the plasma adiponectin levels and that we have to take the CDH13 genotype into account before considering the functional relevance of the adiponectin level. © 2011 Wiley Periodicals, Inc.

Background: The National Nutritional Survey in Japan (NNSJ) was initiated in 1946. Using the majority of the participants for NNSJ, the National Survey on Circulatory Disorders has been conducted every 10 year since 1960. We performed a comparative study of the NNSJ80 estimated individual-based nutritional data by comparing those with NIPPON DATA80 food frequency questionnaires (FFQ) data. Methods: A total of 10 546 community residents from 300 randomly selected districts participated in the both surveys in 1980. At baseline, history, physical, and blood biochemical measurement and a nutritional survey by FFQ were performed individually. From household-based NNSJ80 data, we estimated nutrient intakes of each household member by dividing household intake data proportionally using average intakes by sex and age groups calculated for NNSJ95. We re-categorized NNSJ80 estimated data to correspond to NIPPON DATA80 FFQ categories. Data were analyzed in men and women separately. Results: Cross tables showed fairly good agreement of the two categories. The majorities of participants situated on the diagonally aligned cells or the next to them. Weighted kappa ranged from 0.152 to 0.241. Spearman's rank correlation coefficients between the two categories ranged from 0.224 to 0.338, and those between NNSJ80 continuous data and NIPPON DATA80 categorical data ranged from 0.237 to 0.354. All these values have P < 0.001. Conclusions: These results may indicate that the present nutritional estimation method is applicable to, further studies. © 2010 by the Japan Epidemiological Association.

Hypertension is one of the most common complex genetic disorders. We have described previously 38 single nucleotide polymorphisms (SNPs) with suggestive association with hypertension in Japanese individuals. In this study we extend our previous findings by analyzing a large sample of Japanese individuals (n=14 105) for the most associated SNPs. We also conducted replication analyses in Japanese of susceptibility loci for hypertension identified recently from genome-wide association studies of European ancestries. Association analysis revealed significant association of the ATP2B1 rs2070759 polymorphism with hypertension (P=5.3×10; allelic odds ratio: 1.17 [95% CI: 1.09 to 1.26]). Additional SNPs in ATP2B1 were subsequently genotyped, and the most significant association was with rs11105378 (odds ratio: 1.31 [95% CI: 1.21 to 1.42]; P=4.1×10). Association of rs11105378 with hypertension was cross-validated by replication analysis with the Global Blood Pressure Genetics consortium data set (odds ratio: 1.13 [95% CI: 1.05 to 1.21]; P=5.9×10). Mean adjusted systolic blood pressure was highly significantly associated with the same SNP in a meta-analysis with individuals of European descent (P=1.4×10). ATP2B1 mRNA expression levels in umbilical artery smooth muscle cells were found to be significantly different among rs11105378 genotypes. Seven SNPs discovered in published genome-wide association studies were also genotyped in the Japanese population. In the combined analysis with replicated 3 genes, FGF5 rs1458038, CYP17A1, rs1004467, and CSK rs1378942, odds ratio of the highest risk group was 2.27 (95% CI: 1.65 to 3.12; P=4.6×10) compared with the lower risk group. In summary, this study confirmed common genetic variation in ATP2B1, as well as FGF5, CYP17A1, and CSK, to be associated with blood pressure levels and risk of hypertension. © 2010 American Heart Association, Inc.

Background: We compare the direct homogeneous low-density lipoprotein cholesterol (LDL-C) assay with the Friedewald formula (FF) for determination of LDL-C in a large community-dwelling population. Methods: A total of 21,194 apparently healthy subjects aged 40 to 79 years with triglyceride (TG) concentrations < 4.52 mmol/l were enrolled. LDL-C were directly measured by the enzymatic homogeneous assay (LDL-C (D)) and also estimated by the FF (LDL-C (F)). Paired t-test, Pearson's correlation coefficient and linear regression analysis were performed and the concordances of the National Cholesterol Education Program (NCEP) risk category were estimated. Results: Both in fasting (n. = 3270) and nonfasting samples (n. = 17,924), LDL-C (D) highly correlated with LDL-C (F): r= 0.971 and 0.955, respectively. Concordant results for NCEP categories were 84.8% for fasting samples and 80.1% for nonfasting samples. However, the bias between the 2 measurements increased in samples with TG concentrations > 1.69 mmol/l, especially in nonfasting samples. Conclusions: The results showing less variability of the direct LDL-C assay than that of the FF in nonfasting samples suggest that epidemiological studies can use LDL-C measured by the direct assay both in fasting and nonfasting samples. © 2010 Elsevier B.V.

This study examines the differences in circulating levels of cytokines among Japanese in Japan (JJ), Japanese Americans (JA), and whites and their associations with obesity and marine n-3 fatty acids (FA) in a cross-sectional population-based study of 297 men aged 40-49 (100 JJ, 99 whites, and 98 JA). Experimental studies show that cytokines are associated with obesity positively and marine n-3 FA inversely. Serum interleukin-1α (IL-1α), IL-1 receptor agonist (IL-1ra), IL-4, IL-8, IL-10, inducible protein-10 (IP-10), tumor necrosis factor-α (TNF-α), monocyte chemoattractant protein-1 (MCP-1), and marine n-3 FA were determined. Body mass index (BMI), waist circumference, and computed tomography-measured visceral and subcutaneous adipose tissues were determined. The JJ had significantly lower levels of IL-1α, IL-4, IL-8, MCP-1, and TNF-α than whites and JA. Whites and JA had similar levels of IL-1α, IL-4, and IL-8 whereas whites had significantly higher levels of MCP-1 and TNF-α than JA. The JJ were least obese (BMI (kg/m2), mean-standard deviation) 23.6-2.8, 27.9-4.6, and 27.9-4.5 for JJ, whites, and JA, respectively. The JJ had marine n-3 FA about 100% higher than whites and JA (serum marine n-3 FA (%), median (interquartile range) 8.79 (7.41, 11.16), 3.47 (2.63, 4.83), and 4.44 (3.33, 6.01) for JJ, whites, and JA, respectively). Generally cytokines had weak and nonsignificant associations with indices of obesity and nonsignificant associations with marine n-3 FA. BMI had significant inverse associations with IL-1α, IL-4, and IL-8 in JA (P?

Objective: To investigate associations between socioeconomic status (SES) and prevalence, awareness, treatment, and control of hypertension in Japan's general population. Methods: In 2010, we established a cohort of participants in the National Health and Nutrition Survey of Japan. The cohort included 2623 adults from 300 randomly selected areas across Japan. Using baseline data, four crosssectional analyses were performed: association of SES with prevalent hypertension in 2623 participants; association of SES with unawareness of hypertension and with no treatment in 1282 hypertensive patients; and association of SES with uncontrolled hypertension in 720 treated hypertensive patients. SES was classified according to employment status, length of education, marital and living status, and household expenditure. Results: The overall prevalence of hypertension was 48.9%. Among hypertensive participants, the proportions of unawareness and no treatment were 33.1 and 43.8%, respectively. Target blood pressure levels were not achieved among 61.2% of treated hypertensive patients. Hypertension was more prevalent in the unmarried and living alone group than in the married group (odds ratio 1.76; 95% confidence interval 1.26-2.44), after adjustment for age, sex, BMI, smoking, alcohol consumption, habitual exercise, history of cardiovascular diseases, diabetes mellitus, hypercholesterolemia, dietary sodium, and potassium intake. SES was not clearly associated with unawareness, no treatment, nor poor control of hypertension. Conclusion: Being unmarried and living alone was associated with increased prevalence of hypertension. There was no clear association of SES with unaware, untreated, and uncontrolled hypertension.

Aim: This study was conducted to examine whether babies born small develop cardiovascular problems in later life. Methods: Data were examined retrospectively for 1241 city dwellers (men 521, women 720) aged 40–69 years who received a medical examination at a single institution. Blood tests, physical measurements, and a questionnaire survey regarding birth weight (small, medium, large), medication history and lifestyle (alcohol consumption, smoking, exercise) was administered. We selected 28 persons (12 men and 16 women) from and studied the rate of conformity between the participants’ memory (small, medium, large) and the actual birth weight recorded in their maternal and child health handbook (< 2500, 2500–3500, > 3500 g). Results: Participants’ recall of their birth weight correlated well with the weight recorded in the maternal notebook (r = 0.73; P < 0.025). Low-density lipoprotein (P < 0.05), and total cholesterol (P < 0.01) levels in men, and systolic (P < 0.05) and diastolic (P < 0.05) blood pressure in women were significantly inversely related to birth weight when controlling for age, body mass index, medication, and lifestyle. The percentage of women born small, medium, and large taking antidiabetic agents was 17.8%, 2.9% and 0% (P < 0.05), respectively. Conclusion: A low birth weight was associated with high low-density lipoprotein and total cholesterol levels in men, and hypertension and diabetes mellitus in women aged 40–69.

Aim: In Japan Atherosclerosis Society guidelines for the prevention of atherosclerotic cardiovascular diseases 2012 (JAS2012), NIPPON DATA80 risk assessment chart (ND80RAC) was adopted to estimate the 10-year probability of coronary artery disease (CAD) mortality. However, there was no comparison between the estimated mortality calculated by ND80RAC and actual mortality in external populations. Accordingly, we used the large pooled database of cohorts in Japan, EPOCH-JAPAN, as an external population.Methods:The participants of EPOCH-JAPAN without a history of cardiovascular disease (15,091 men and 18,589 women aged 40–74 years) were analyzed based on sex. The probability of a 10-year risk of CAD/stroke mortality was estimated by ND80RAC. The participants were divided into both decile of their estimated mortality and three categories according to JAS2012. The calibration between the mean estimated mortality and the actual mortality was performed by the Hosmer and Lemeshow (H-L) test.Results: In both sexes, the estimated CAD mortality was higher than the actual mortality, particularly in higher deciles of estimated mortality, and the estimated stroke mortality was almost concordant with the actual mortality in low/moderate deciles of estimated mortality. As for the categories according to JAS2012, the estimated CAD mortality was higher than the actual mortality in both sexes; actual mortality in Category III was lower than that in Category II in women. However, it increased in the ascending order of category when we excluded the presence of diabetes from Category III.Conclusions: The estimated CAD mortality by ND80RAC tended to be higher than the actual mortality in the population in which the baseline survey was more recently performed.

Backgrounds: The association between weight gain and the incidence of type 2 diabetes is well known. The aim of our study was to investigate the relationship between change in waist circumference (WC) and type 2 diabetes incidence. Methods: The participants in the Suita Study, a population-based cohort study in an urban area of Japan, underwent a baseline survey between 1989 and 1994 (Exam 1) and were examined at follow-up every 2 years. We performed a 9.3-year cohort study of 946 men and 1327 women with no history of diabetes who underwent Exam 1 and Exam 2 (between 1997 and 1999). Participants were stratified by sex and median WC at Exam 1, and, in each stratum, participants were further classified into three categories by tertile of WC change per year between Exam 1 and Exam 2. Hazard ratios (HRs) and 95% confidence intervals (CIs) for type 2 diabetes incidence were calculated by Cox proportional hazard models. The endpoints were first diagnosis of type 2 diabetes or March 2011. Results: During follow-up, 287 participants developed type 2 diabetes. In both sexes with median WC or higher, participants in the highest tertile of WC change had a significantly higher risk of developing type 2 diabetes. Multivariable adjusted HRs were 1.84 (95% CI, 1.10-3.08) in men and 2.30 (95% CI, 1.31-4.04) in women. No significant association was observed among participants with WC below median. Conclusions: Preventing WC gain is important in preventing type 2 diabetes in the Japanese population, especially among individuals with a relatively high WC.