Published papers

Among Asians, including Japanese, obesity is related to dyslipidemia and insulin resistance at a lower level of body mass index (BMI) compared with non-Hispanic whites (NHW). We hypothesize that this ethnic difference in the relationship between BMI and metabolic risks is partly associated with the ethnic difference in fat distribution, namely, liver fat as well as visceral adipose tissue. To compare liver fat content among Japanese vs NHW men, regional computed tomographic images were taken to measure liver computed tomographic density in population-based samples of 313 Japanese and 288 NHW men aged 40 to 49 years, along with the assessment of metabolic parameters. Liver fat content was higher in Japanese than NHW men (liver to spleen attenuation ratio [lower value means higher liver fat content]: 1.01 ± 0.16 vs 1.07 ± 0.15, respectively; P < .01), despite a lower mean BMI in Japanese men (BMI: 23.6 ± 2.9 vs 27.8 ± 4.2 kg/m2, P < .01). Moreover, Japanese men had a greater disposition for fatty liver with a small increase in BMI than NHW (P < .01), whereas both groups had a similar relationship between visceral adipose tissue and BMI. In both groups, liver fat content correlated with triglycerides, homeostasis model assessment of insulin resistance, and C-reactive protein. Liver fat content is higher among Japanese than NHW; and this ethnic difference becomes more robust with a small increase in BMI, suggesting that fatty liver is a sensitive marker for the failure of the adipose tissue to expand to accommodate an increased energy influx, and is associated with similar metabolic risk in Japanese despite lower BMI compared with NHW men. © 2009 Elsevier Inc. All rights reserved.

We examined whether the A1166C polymorphism of the angiotensin II type I receptor gene (AGTR1) affects cardiovascular event occurrence in a Japanese prospective cohort study. The 2212 participants who gave informed consent for genetic analysis were enrolled in this study (the Shigaraki Study). The average observation period was 1954 days. Cardiovascular events occurred in 37 individuals (1.7%). The independent factors which specified cardiovascular events were age (hazard ratio (HR) = 1.13; 95% confidence interval (CI): 1.10-1.16; p < 0.0001) and sex (HR = 2.18; 95%CI: 1.23-3.85; p = 0.007). However, the A1166C polymorphism of AGTR1 was not a predictive factor for cardiovascular events (HR = 1.11; 95%CI: 0.61-2.02; p = 0.731). © 2008 Elsevier Ireland Ltd. All rights reserved.

Background: The association of serum polyunsaturated fatty acids (PUFAs) with lipids in different populations is not known. Objective: Our aim was to examine the association of serum n-6 (omega-6) or n-3 (omega-3) PUFAs with triglycerides or HDL-cholesterol concentrations in 261 white, 285 Japanese, and 212 Japanese American men aged 40-49 y. Design: We used a population-based cross-sectional study. Of the original sample (n = 926), those taking lipid-lowering medications or who had diabetes (n = 168) were excluded. Serum fatty acids were analyzed by capillary gas-liquid chromatography. Multiple regression models as a function of tertile groups of each PUFA were used. Results: Serum n = 6 PUFAs were significantly inversely associated with triglycerides across populations after adjustment for age, body mass index, pack-years of smoking, and ethanol consumption [β = -0.39 (P < 0.001), -0.38 (P < 0.001), and -0.33 (P < 0.001) in whites, Japanese, and Japanese Americans, respectively]. Marine n-3 PUFAs were significantly inversely associated with triglycerides across populations [β = -0.15 (P < 0.001), -0.22 (P < 0.001), and -0.13 (P < 0.001) in whites, Japanese, and Japanese Americans, respectively]. n-6 PUFAs were significantly positively associated with HDL cholesterol in whites (β = 4.49, P < 0.001) and Japanese (β = 3.73, P < 0.01). Marine n-3 PUFAs were significantly positively associated with HDL cholesterol in Japanese (β = 2.15, P < 0.05), and eicosapentaenoic acid was significantly positively associated with HDL cholesterol in whites (β = 2.68, P < 0.01). Conclusion: Serum n-6 and n-3 PUFAs are inversely associated with triglycerides across populations. © 2009 American Society for Nutrition.

Background: Recent genome-wide association studies (GWAS) have identified genes or loci affecting lipid levels. Given the difference in allele frequencies and linkage disequilibrium patterns across the populations, a GWAS was conducted using the Illumina 550K in a Japanese population (n=900) in search of population-specific genetic variations associated with high-density lipoprotein (HDL)-cholesterol. Methods and Results: Among the 368,274 single nucleotide polymorphisms (SNPs) with a minor allele frequency of at least 0.1, 43 SNPs exceeded the arbitrary threshold of -log10P >4.0. The most significant SNP was rs3764261, located 5′upstream of CETP, exhibiting a -log10P value of 6.17. Increasing the sample size by genotyping in the additional Suita sample (n=1,810) further improved the level of significance, with each additional copy of the minor allele being associated with an increase in HDL-cholesterol by 6.2 mg/dl (P=3.4×10-12). Interestingly, the minor allele was more prevalent in cases with myocardial infarction than in controls (0.221 vs 0.196, nominal P=0.02). Conclusions: The association between genetic variants at CETP and HDL-cholesterol was replicated in our sample. None of the genetic variants exerted a greater influence on HDL levels than those at CETP. Associations for the top-ranked SNPs need to be tested for further replication in an independent sample.

Background: In myocardial infarction (MI), it is well known that serum creatine kinase (s-CK) increases after onset, but it is unclear whether s-CK elevates before MI onset. The present analysis examined whether elevated s-CK levels predicted first-ever MI or stroke. Methods: This study was a population-based cohort study in a Japanese urban area. Study subjects were comprised of 5026 initially healthy Japanese (2370 men and 2656 women, mean age: 54.5 years) without a history of MI or stroke. They were followed-up for 11.8 years on average, and 103 MIs (definite: 45; probable: 58) and 168 strokes (definite: 126; probable: 42) were observed. There was no subject who developed MI just at baseline (the follow-up period among those with definite MI was, at earliest, 0.20 years). Results: The adjusted hazard ratio for definite MI was 4.18 (95% confidence interval 1.66-10.53) with s-CK levels of ≥200 IU/l, compared with the reference category (s-CK levels of ≤99 IU/l), whereas no relationship was observed between s-CK levels and the risk for stroke. With regard to definite MI, an interaction between s-CK levels and dyslipidaemia was observed. Among subjects with hypercholesterolaemia, the hazard ratio linearly elevated with increased s-CK levels. On the other hand, no linear elevation was observed among subjects without hypercholesterolaemia (P for interaction = 0.011). Conclusions: The present study suggested that screening for elevated s-CK levels in initially healthy Japanese subjects was useful to predict first-ever MI in the future, especially in subjects with dyslipidaemia. Published by Oxford University Press on behalf of the International Epidemiological Association. © The Author 2009.

Objective: Only a small number of population-based cohort studies have directly compared the predictive value of low-density lipoprotein cholesterol (LDL-C) and non-high-density lipoprotein cholesterol (non-HDLC) for coronary artery disease in Asian populations, such as Japan. Methods: We performed an 11.9-year cohort study of 4694 men and women, aged 30-74 years, selected randomly from an urban general population in Japan. Baseline LDL-C levels were estimated using the Friedewald formula. The predictive values of LDL-C and non-HDLC for myocardial infarction (MI) and stroke were compared. Results and conclusion: During the follow-up period, there were 80 incident cases of MI and 139 of stoke, comprised of 23 intracerebral hemorrhages, 85 cerebral infarctions and 31 other types of stroke. The Hazard ratio (HR) for MI was highest in the top quintile of LDL-C (HR: 3.03, 95% CI, 1.32-6.96) when male and female data were combined. The HR for MI was also highest in the top quintile of non-HDLC (HR: 2.97, 95% CI, 1.26-6.97). Analysis of trends showed a significant positive relationship between MI incidence and serum LDL-C and non-HDLC levels (both P = 0.02). However, there was no relationship between the incidence of any subtype of stroke and either LDL-C or non-HDLC. The predictive value of LDL-C and non-HDLC for MI, assessed by calculating the differences in the -2 logarithm likelihood (-2 ln [L]) and area under the curve (AUC), were almost similar. © 2008 Elsevier Ireland Ltd. All rights reserved.

We sought to examine effects of habitual alcohol intake on ambulatory blood pressure (BP), heart rate (HR), and HR variability among Japanese men. Subjects were 539 men aged 35 to 65 years from rural and urban communities. Ambulatory BP and HR were monitored with an automated, portable, noninvasive multibiomedical recorder. Power spectral analysis of the RR intervals on the ECG was performed every 5 minutes. Compared with nondrinkers, moderate drinkers (alcohol intake 23 to 45 g/d) and heavy drinkers (alcohol intake ≥46 g/d) showed higher age- and field-adjusted mean values of systolic and diastolic BPs during the morning and while awake, but there were no differences in BPs over 24-hour periods and while asleep among the alcohol intake categories. Alcohol intake was positively associated with mean values of sleep-morning differences and daytime variability in BPs, HRs while awake and asleep, and low frequency: high frequency ratio while asleep. The results were virtually unchanged after adjustment for body mass index, smoking, and diabetes mellitus. Compared with the nondrinkers, age- and field-adjusted odds ratios of the morning BP surge (excess elevation of BP in the morning: morning systolic BP minus sleep systolic BP ≥37 mm Hg) for light (alcohol intake 0 to 22 g/d), moderate, and heavy drinkers were 0.96 (95% CI: 0.34 to 2.78), 1.68 (95% CI: 0.64 to 4.38), and 2.73 (95% CI: 1.12 to 6.67), respectively. Habitual alcohol intake was associated with increased BP in the morning, HR while awake and asleep, and sympathetic activity while asleep, which may explain some of the mechanisms of the relationship between heavy alcohol intake and risk of cardiovascular diseases. © American Heart Association, Inc.

Tissue factor pathway inhibitor (TFPI) is an anticoagulant protease inhibitor that inhibits the tissue factor-initiated blood coagulation cascade reactions. Based on these anticoagulant functions of TFPI, we hypothesized that genetic variations in TFPI may alter the TFPI expression or impair the anticoagulant function and could predispose persons to deep vein thrombosis (DVT). This study was undertaken to examine whether the genetic polymorphisms in TFPI are associated with the plasma TFPI levels and risk for DVT. We sequenced the entire coding regions of TFPI in 175 Japanese DVT patients and identified 12 genetic variants, including one missense mutation, Asn221Ser. The missense mutation occurred at the site presumably attached to the glycosylphosphatidylinositol anchor in the TFPI-p form. The allele frequency of the mutant Ser-coding allele of the Asn221Ser mutation was 8% in the Japanese general population consisting of 1684 individuals. The Asn221Ser mutation was significantly associated with the total TFPI levels (Asn/Asn, n = 108, total TFPI = 56.57 ± 0.88 ng/ml (mean ± SD) vs. Asn/Ser + Ser/Ser, n = 16, total TFPI = 63.44 ±.28 ng/ml, P= 0.0058). The genotype was not associated with the free TFPI level. This Asn221Ser mutation was not associated with DVT. Thus, the Asn221Ser mutation occurring in the TFPI-p form was associated with the total TFPI level, but not a risk for DVT. The absence of the putative glycosylphosphatidylinositol anchor in TFPI-β under pathological conditions remains to be studied. Blood Coagul Fibrinolysis 20:22-26 © 2009 Wolters Kluwer Health Lippincott Williams & Wilkins.

Excess cardiovascular risk in men compared with women has been suggested to be partly explained by effects of the Y chromosome. However, inconsistent results have been reported on the Y chromosome's genetic influence on blood pressure and lipid levels. The purpose of the present study was to settle the question whether genetic variants of the Y chromosome influence cardiovascular risk factors using a large epidemiological cohort, the Suita study. Possible influences of the Y chromosome polymorphisms (Y chromosome Alu insertion polymorphism [YAP], M175 and SRY+465) on cardiovascular risk factors were assessed in 974 Japanese men. The frequency of the YAP(+) allele in our study sample was 0.31. The prevalence of hypertension tended to be higher in YAP(+) than in YAP(-) men, and this tendency was found to be stronger among men aged 65 years or older. Men with the YAP(+) genotype had higher levels of high density lipoprotein (HDL) cholesterol compared with those with the YAP(-) genotype, even after adjustment for age, body mass index, and daily ethanol and cigarette consumption (57.0±14.6 mg/dL vs. 54.2±14.2 mg/dL, nominal p=0.011, adjusted p=0.0062). However, these observed nominal associations disappeared after adjusting for multiple testing (Bonferroni). No association was detected between the YAP genotype and myocardial infarction. Similarly, none of the associations with M175 and SRY+465 attained significance when multiple testing was taken into account. In conclusion, Y chromosome polymorphisms (YAP, M175 and SRY+465) do not appear to be associated with cardiovascular risk factors in Japanese men. Studies using much larger sample sizes and/or additional independent samples will be required for definitive conclusions.

Background: Some studies focusing on low-risk profiles for cardiovascular disease have been reported in Western countries. Yet, few reports have examined, with substantial longevity, the low-risk profile for cardiovascular disease in the Japanese population. This study examines whether having a favorable risk factor profile yields lower all-cause mortality and whether the proportion of those with a low-risk profile is larger in the Japanese population. Methods and Results: A total of 8,339 men and women aged 30-69 years without a history of cardiovascular diseases for 19 years, who had participated in the 1980 National Survey on Circulatory Disorders after being randomly selected from throughout Japan, were followed. Low risk was defined as having all of the following baseline characteristics: blood pressure (BP) <120/80 mmHg; no antihypertensive medication; serum cholesterol 160-240 mg/dl (4.14-6.22 mmol/L); no history of diabetes; and non-smoker. The long-term mortality of the low-risk group was compared with that of others using the Cox proportional hazard model. The prevalence of low risk was 9.4% of all participants. The multivariate-adjusted hazard ratios for low-risk individuals compared with others were as follows: 0.33 (95% confidence intervals (CI), 0.15-0.74) for cardiovascular disease and 0.63 (95% CI, 0.46-0.88) for all-cause mortality. The most attributable risk factor for all-cause mortality was high BP (≥120/80 mmHg). Conclusion Japanese individuals with favorable cardiovascular disease risk profiles had lower mortality from cardiovascular disease and all-causes than those without.

Abdominal obesity is a prerequisite for some definitions of metabolic syndrome (MetS). We investigated the impact of MetS defined by two different criteria, which either did or did not require abdominal obesity as a prerequisite, on cardiovascular disease (CVD) incidence in an urban Japanese cohort study. We studied 5,332 Japanese (aged 30-79 years, without CVD at baseline), who completed a baseline survey (September 1989 to March 1994) and were followed up through December 2005. MetS was defined by the NCEP-ATPIII (modified by Asian obesity criteria) and the Japanese criteria. After 61,846 person-years of follow-up, we documented 317 CVD incidences. The MetS frequencies of the Japanese and of the modified NCEP-ATPIII criteria were 17.7% and 25.1% for men and 5.0% and 14.3% for women, respectively. The multivariate hazard ratios (HRs; 95% confidence intervals [CI]) of CVD incidence for MetS by the modified NCEP-ATPIII criteria were 1.75 (1.27-2.41) in men and 1.90 (1.31-2.77) in women, and those for MetS by the Japanese criteria were 1.34 (0.96-1.87) in men and 2.20 (1.31-3.68) in women. The multivariate HRs of CVD incidence for MetS for the Japanese and for the modified NCEP-ATPIII criteria were 2.92 (1.54-5.55) and 1.94 (0.98-3.82) in men under 60 years old, respectively. The CVD incidence risks increased according to the number of MetS components. The risks were similar among participants with the same number of MetS components, regardless of abdominal obesity. In conclusion, the number of MetS components (modified NCEP-ATPIII criteria) may be more strongly associated with CVD incidence than the abdominal obesity essential criteria (the Japanese criteria) in a general urban Japanese population.

A family history of stroke seems to be related with increased risk of stroke although the relationship is not always significant. Increased risk of stroke is strongly associated with hypertension, which might be also associated with family history. However, investigations into the relationship between family history of hypertension and stroke mortality are scarce. We investigated whether a family history of stroke and that of hypertension evaluated using a simple questionnaire could predict stroke mortality in Japanese. We obtained parental histories of stroke and of hypertension from 8,037 randomly selected general Japanese without history of cardiovascular disease and followed them for 19 years. The multivariate adjusted hazard ratios (HRs) for total stroke mortality, intra-cerebral hemorrhage mortality and for cerebral infarction mortality according to family history were estimated using the Cox proportional hazards model. The prevalences of family histories of stroke and of hypertension were 20.6% and 31.1%, respectively. A family history of stroke was not related to total stroke mortality, intra-cerebral hemorrhage mortality or to cerebral infarction mortality. Meanwhile, a family history of hypertension was positively related to total stroke mortality among women aged less than 60 years and men aged 60 or more years (women: HR=3.41, 95% confidence interval [CI]: 1.49-7.81; men: HR=1.50, 95% CI: 1.00-2.24) even after adjustment for systolic blood pressure. In conclusion, a family history of stroke could not predict total stroke mortality. However, a family history of hypertension might predict an increased risk for total stroke.

purpose: We sought to identify factors related to total mortality in older Japanese men in Japan and Hawaii. Methods: Baseline data were collected from 1980 to 1982 in 1379 men in Hawaii and 954 men in Japan. Ages ranged from 61 to 81 years, with mortality follow-up during a 19-year period. Results: Compared with Japan, men in Hawaii had a 2-fold excess of diabetes and a 4-fold excess of prevalent coronary heart disease (P < .001). Total cholesterol and body mass index were also greater in Hawaiian men (P < .001). In contrast, men in Japan had greater systolic blood pressure and were nearly 3 times more likely to smoke cigarettes (P < .001). Although each cohort had elements of a poor risk factor profile, there was a 1.4-fold excess in the risk of death in Japan (49.4 vs. 36.2/1,000 person-years, P < .001). Although mortality was similar after risk factor adjustment, only blood pressure and cigarette smoking accounted for the higher risk of death in Japan. Conclusions: Cigarette smoking and hypertension explain much of the excess mortality in Japan versus Hawaii. In this comparison of genetically similar cohorts, evidence further suggests that Japanese in Japan are equally susceptible to develop the same adverse risk factor conditions that exist in Hawaii. © 2008 Elsevier Inc. All rights reserved.

Few prospective studies have examined the association between high-normal blood pressure and cardiovascular disease (CVD) in Asia. We examined the impact of high-normal blood pressure on the incidence of CVD in a general urban population cohort in Japan. We studied 5494 Japanese individuals (ages 30 to 79 years without CVD at baseline) after completing a baseline survey who received follow-up through December 2005. Blood pressure categories were defined on the basis of the ESH-ESC 2007 criteria. In 64 391 person-years of follow-up, we documented the incidence of 346 CVD events. The frequencies of high-normal blood pressure and hypertension Stage 1 and Stage ≥2 were 18.0%, 20.1%, and 10.1% for men and 15.9%, 15.6%, and 8.8% for women, respectively. Antihypertensive drug users were also classified into the baseline blood pressure categories. Compared with the optimal blood pressure group, the multivariable hazard ratios (95% confidence intervals) of CVD for normal and high-normal blood pressure and hypertension Stage 1 and Stage ≥2 were 2.04 (1.19 to 3.48), 2.46 (1.46 to 4.14), 2.62 (1.59 to 4.32), and 3.95 (2.37 to 6.58) in men and 1.12 (0.59 to 2.13), 1.54 (0.85 to 2.78), 1.35 (0.75 to 2.43), and 2.86 (1.60 to 5.12) in women, respectively. The risks of myocardial infarction and stroke for each blood pressure category were similar to those of CVD. Population-attributable fractions of high-normal blood pressure and hypertension for CVD were 12.2% and 35.3% in men and 7.1% and 23.4% in women, respectively. In conclusion, high-normal blood pressure is a risk factor for the incidence of stroke and myocardial infarction in a general urban population of Japanese men. © 2008 American Heart Association, Inc.

Objectives: We sought to examine whether marine-derived n-3 fatty acids are associated with less atherosclerosis in Japanese versus white populations in the U.S. Background: Marine-derived n-3 fatty acids at low levels are cardioprotective through their antiarrhythmic effect. Methods: A population-based cross-sectional study in 281 Japanese (defined as born and living in Japan), 306 white (defined as white men born and living in the U.S.), and 281 Japanese-American men (defined as Japanese men born and living in the U.S.) ages 40 to 49 years was conducted to assess intima-media thickness (IMT) of the carotid artery, coronary artery calcification (CAC), and serum fatty acids. Results: Japanese men had the lowest levels of atherosclerosis, whereas whites and Japanese Americans had similar levels. Japanese had 2-fold higher levels of marine-derived n-3 fatty acids than whites and Japanese Americans in the U.S. Japanese had significant and nonsignificant inverse associations of marine-derived n-3 fatty acids with IMT and CAC prevalence, respectively. The significant inverse association with IMT remained after adjusting for traditional cardiovascular risk factors. Neither whites nor Japanese Americans had such associations. Significant differences between Japanese and whites in multivariable-adjusted IMT (mean difference 39 μm, 95% confidence interval [CI]: 21 to 57μm, p < 0.001) and CAC prevalence (mean difference 10.7%, 95% CI: 2.9% to 18.4%, p = 0.007) became nonsignificant after we adjusted further for marine-derived n-3 fatty acids (22 μm, 95% CI: -1 to 46 μm, p = 0.065 and 5.0%, 95% CI: -5.3% to 15.4%, p = 0.341, respectively). Conclusions: Very high levels of marine-derived n-3 fatty acids have antiatherogenic properties that are independent of traditional cardiovascular risk factors and may contribute to lower the burden of atherosclerosis in Japanese, a lower burden that is unlikely the result of genetic factors. © 2008 American College of Cardiology Foundation.