Laura Newell-Morris

Visceral Adiposity Is an Independent Predictor of Incident Hypertension in Japanese Americans

Context Central obesity and hypertension are well-established components of the metabolic syndrome, but what exactly is the relationship between visceral adiposity and hypertension? Contribution This prospective study used computed tomography to measure multiple body fat areas of 300 middle-aged, normotensive Japanese Americans. Ninety-two participants developed hypertension within 10 to 11 years. Greater visceral adiposity, independent of other measures of body fat and other risk factors, such as plasma insulin and glucose levels, was associated with increased risk for hypertension. Cautions Relationships between visceral adiposity and the development of hypertension may vary in different ethnic groups. The Editors A central pattern of body fat distribution is now generally considered to play an important role in the metabolic syndrome, which involves obesity, insulin resistance, hyperinsulinemia, dyslipidemia, glucose intolerance, and hypertension (1, 2). In particular, visceral adiposity rather than regional or generalized obesity appears to play a key role in these diseases (3-7). Several cross-sectional and prospective studies have examined associations between hypertension and greater central obesity, as measured by waist circumference, the ratio of waist-to-hip circumference, or the ratio of subscapular-to-triceps skinfold thickness (8-15). The cross-sectional studies have reported a positive association (8, 9), but the prospective studies have been inconclusive (10-15). These studies have posited that visceral adiposity and insulin resistance are the most important factors linking greater abdominal obesity (as assessed by surface measurements) and hypertension. Although visceral fat is thought to affect the prevalence of hypertension, only 3 cross-sectional studies have suggested a possible association between visceral adiposity (measured by using computed tomography [CT]) and blood pressure (3, 4, 16); however, the results of these studies were inconclusive. No prospective studies have examined whether directly measured visceral fat is associated with an increased risk for incident hypertension. Therefore, we prospectively examined the relationship between directly measured visceral adiposity and the risk for incident hypertension, independent of other measurements of total and regional adiposity and fasting plasma insulin. Methods Study Sample Between 1983 and 1988, we enrolled 658 second- and third-generation Japanese Americans who were between 34 and 76 years of age (mean age, 54.2 years) into the Japanese American Community Diabetes Study (17, 18). Participants were chosen from volunteers through community-wide recruitment and were representative of Japanese-American residents of King County, Washington, in age distribution, residential distribution, and parental immigration pattern. A comprehensive mailing list and telephone directory that included almost 95% of the Japanese-American population of King County, Washington, was used. All participants were of 100% Japanese ancestry. Participants returned for follow-up examinations 5 to 6 and 10 to 11 years after a baseline evaluation. For the current analysis, eligible participants had systolic blood pressure less than 140 mm Hg and diastolic blood pressure less than 90 mm Hg and were not taking antihypertensive or oral hypoglycemic medications or insulin. We excluded 277 of the 658 participants in the original cohort because they did not meet the inclusion criteria. We excluded an additional 67 persons because of death, loss to follow-up, or withdrawal from the study. Another 14 persons who completed follow-up but had missing covariate information were also excluded. The analytic cohort consisted of 300 persons (Figure). The follow-up rate in the present study was 91% (345 of 381) at the 5- to 6-year examination and 80% (304 of 381) at the 10- to 11-year examination (Figure). Figure. Flow of participants through the study. Data Collection All measurements were made in the General Clinical Research Center at the University of Washington, Seattle, Washington. The Human Subjects Review Committee at the University of Washington approved the protocol for this research, and we obtained signed informed consent from all participants. At all examinations, blood pressure was measured to the nearest 2 mm Hg with a mercury sphygmomanometer while the participant was in a recumbent position. Systolic blood pressure was determined by the first perception of sound, and diastolic blood pressure was determined at the disappearance of sounds (fifth-phase Korotkoff). Average blood pressure was calculated from the second and third of 3 consecutive measurements. We diagnosed hypertension at baseline or follow-up if the average systolic blood pressure was 140 mm Hg or greater, the average diastolic blood pressure was 90 mm Hg or greater, or the participant was receiving antihypertensive medications. We classified participants as hypertensive if they met these criteria at the follow-up examination at 5 to 6 years or 10 to 11 years (Figure). All patients received a 75-g oral glucose tolerance test after a 10-hour fast. We then used the American Diabetes Association criteria (19) to classify patients as having normal glucose tolerance, impaired glucose tolerance, or type 2 diabetes mellitus. Blood samples were drawn after an overnight 10-hour fast and during an oral glucose tolerance test for measurement of plasma glucose and insulin levels. We used an automated glucose oxidase method to assay plasma glucose. Fasting plasma insulin was measured by radioimmunoassay, as reported previously (5, 7). We measured triglyceride and high-density lipoprotein cholesterol levels in the Northwest Lipid Research Laboratory, according to modified procedures of the Lipid Research Clinics (20). We calculated body mass index (BMI) as the weight in kilograms divided by height in meters squared. For CT scans, we used single slices of the thorax, abdomen (at the umbilicus), and mid-thigh to measure cross-sectional subcutaneous thoracic, abdominal, and right thigh and intra-abdominal fat areas (measured in cm2), as described elsewhere (21). We directly estimated visceral adiposity from the intra-abdominal fat area. This measurement has been reported to have a high correlation with directly ascertained total visceral fat volume measured by using CT or magnetic resonance imaging (22, 23). We calculated total subcutaneous fat area as the sum of subcutaneous thoracic and abdominal fat areas and twice the right thigh subcutaneous fat area. We defined total fat area as total subcutaneous fat area plus intra-abdominal fat area. Among Japanese Americans, total fat area correlates highly with fat mass, as measured by hydrodensitometry (r= 0.89 to 0.94) (24). Waist circumference was measured at the level of the umbilicus to the nearest tenth of a centimeter. Participants were questioned about current use of cigarettes and daily consumption of alcoholic beverages, which was converted into grams of alcohol consumed per day. Usual weekly energy expenditure in kilocalories was estimated from questionnaire data on work and recreational activities, strenuous exercise, distance walked, and stairs climbed, as described elsewhere (25). Statistical Analysis We used multiple logistic regression analysis to estimate the odds ratio for incident hypertension in relation to an increase of 1 SD in baseline variables. For rare outcomes, the odds ratio will approximately equal the relative risk. For more frequent outcomes, such as hypertension, the odds ratio will overestimate the relative risk (26). We evaluated nonlinear effects of continuous independent variables by using quadratic and log transformations (27). The linear trends in odds were evaluated by using the median value for each quartile category of continuous variables. To assess departure from linearity, we included linear and quadratic terms (the median and the value squared) in the model (28). To determine whether interaction was present (that is, the relationship between the risk factor and the outcome varied depending on the value of a third variable) (27, 29, 30), we inserted first-order interaction terms into appropriate regression models. We assessed interaction to determine whether the relationship between hypertension status at follow-up and baseline adipose variables, such as intra-abdominal fat area, subcutaneous abdominal fat area, total subcutaneous fat area, BMI, or waist circumference, differed according to the level of an additional variable (for example, sex) in the model. We used the likelihood ratio test to determine the statistical significance of nonlinear effects of continuous independent variables and interaction terms in the logistic regression models. Multicollinearity was assessed by using the variance inflation factor (31). A variance inflation factor exceeding 10 is regarded as indicating serious multicollinearity, and values greater than 4.0 may be a cause for concern (31). We calculated the 95% CI for each odds ratio. P values were 2-tailed. We performed statistical analyses using Stata SE, version 8.0 (Stata Corp., College Station, Texas). Role of the Funding Sources The funding sources had no role in the collection, analysis, or interpretation of the data or in the decision to submit the manuscript for publication. Results Among the 300 eligible men and women followed for 10 to 11 years, there were 92 incident cases of hypertension. In univariate logistic regression analysis, intra-abdominal fat area, abdominal subcutaneous fat area, total subcutaneous fat area, total fat area, BMI, and waist circumference were associated with a higher incidence of hypertension. Fasting plasma insulin level, fasting plasma glucose level, 2-hour plasma glucose level, and high-density lipoprotein cholesterol level were also associated with incidence of hypertension (Table 1). Wealso compared the baseline characteristics of participants included in

Association of Elevated Fasting C-Peptide Level and Increased Intra-Abdominal Fat Distribution With Development of NIDDM in Japanese-American Men

The Japanese-American population of King County, Washington, is known to have a high prevalence of non-insulin-dependent diabetes mellitus (NIDDM). As part of a community-based study, we reexamined 146 second-generation Japanese-American men who had been initially classified as nondiabetic. At a mean follow-up period of 30 mo, 15 men had developed NIDDM, and 131 remained nondiabetic. The variables measured at the initial visit that distinguished the 15 diabetic men from the 131 nondiabetic men were older age, higher serum glucose level at 2 h after 75 g oral glucose, higher fasting plasma C-peptide level, and increased cross-sectional intra-abdominal fat area as determined by computed tomography. Both older age and higher 2-h glucose levels are variables that have been associated with the development of NIDDM, but the association of higher fasting C-peptide level and greater intra-abdominal fat area with subsequent development of NIDDM were new observations. The elevated fasting C-peptide level persisted after adjustment for fasting serum glucose. The elevated C-peptide level represents hypersecretion of insulin and was interpreted to reflect a compensatory response to an underlying insulin-resistant state that antedates the development of NIDDM. The fasting C-peptide level was correlated with the intra-abdominal fat area, suggesting that the intra-abdominal fat area may be associated with insulin resistance. Thus, in individuals who develop NIDDM, insulin resistance, increased insulin secretion, and increased intra-abdominal fat are present before diabetic glucose tolerance can be demonstrated.

Prevalence of diabetes mellitus and impaired glucose tolerance among second-generation Japanese-American men.

We describe the initial findings from a multidisciplinary, epidemiologic study of diabetes mellitus conducted in a population of secondgeneration Japanese-American (Nisei) men born between 1910 and 1939 who reside in King County, Washington (n = 1746). From this study population, 487 volunteered, and 229 were enrolled to comprise the study sample. A random sample of Nisei men was also drawn from the population to develop a reference sample of 189 men. All subjects participated in a 75-g oral glucose tolerance test; the National Diabetes Data Group (NDDG) and World Health Organization (WHO) diagnostic criteria as well as a modification of the WHO criteria were used to classify individuals with normal glucose tolerance, impaired glucose tolerance (IGT), or diabetes. Within the study sample, 79 men were found to have normal glucose tolerance, 72 had IGT, and 78 had type II diabetes. The mean age of the study sample was 61.4 yr. Based on comparison of the study sample to the reference sample, the study sample was ascertained to be representative of Nisei men in King County. Extrapolating from our observations in the reference sample and in the study sample, we have estimated that ∼56% of Nisei men in the study population have abnormal glucose tolerance. Much of this is undiagnosed because only ∼13% of the reference sample of Nisei men reported a prior diagnosis of diabetes. Of the men who enrolled in the study as nondiabetic subjects, 11.1% had diabetes and 39.2% had IGT; i.e., 50.3% had previously unknown abnormalities in glucose tolerance. We estimate that ∼20% of Nisei men have diabetes (both previously diagnosed and undiagnosed) and ∼36% have IGT.

Earlier appearance of impaired insulin secretion than of visceral adiposity in the pathogenesis of NIDDM. 5-Year follow-up of initially nondiabetic Japanese-American men.

OBJECTIVE To identify risk factors for development of non-insulin-dependent diabetes mellitus (NIDDM) during a 5-year longitudinal follow-up of second-generation Japanese-American (Nisei) men. RESEARCH DESIGN AND METHODS For 5 years, 137 initially nondiabetic Nisei men were followed with 75-g oral glucose tolerance tests at the initial visit and at 2.5- and 5-year follow-up visits. Body fat distribution was assessed by computed tomography (CT) and body mass index (BMI) calculated at each visit. Fasting insulin and C-peptide, the increment of insulin and C-peptide at 30 min after the oral glucose load, intra-abdominal and total subcutaneous fat by CT, and BMI were compared between those who remained nondiabetic (non-DM) and those who had developed NIDDM at 2.5 years (DM-A) and 5 years (DM-B). RESULTS At baseline, the DM-A group had significantly increased intra-abdominal fat, elevated fasting plasma C-peptide, and lower C-peptide response at 30 min after oral glucose. At the 2.5-year follow-up, this group had markedly increased fasting plasma insulin and decreased 30-min insulin and C-peptide response to oral glucose. The DM-B group also had significantly lower insulin response at 30 min after oral glucose at baseline but no significant difference in intra-abdominal fat or fasting plasma insulin and C-peptide levels. When this group developed NIDDM by 5-year follow-up, however, an increase of intra-abdominal fat was found superimposed on the pre-existing lower insulin response. Fasting plasma insulin and C-peptide remained low. CONCLUSION In DM-A, lower 30-min insulin response to oral glucose (an indicator of β-cell lesion) and increased intra-abdominal fat and fasting C-peptide (indicators of insulin resistance) were the risk factors related to the development of NIDDM. DM-B subjects had a lower 30-min insulin response to oral glucose at baseline and increased intra-abdominal fat at 5-years, when they were found to have NIDDM. Thus, both insulin resistance and impaired β-cell function contribute to the development of NIDDM in Japanese-Americans, and impaired β-cell function may be present earlier than visceral adiposity in some who subsequently develop NIDDM.

Visceral Adiposity and the Prevalence of Hypertension in Japanese Americans

Background—Visceral adiposity is generally considered to play a key role in the metabolic syndrome, including hypertension. The purpose of this study was to evaluate cross-sectionally whether visceral adiposity is associated with prevalence of hypertension independent of other adipose depots and fasting plasma insulin. Methods and Results—Study subjects included 563 Japanese Americans with normal or impaired glucose tolerance or diabetes but not taking oral hypoglycemic medication or insulin at entry. Variables included plasma glucose and insulin measured after an overnight fast and during an oral glucose tolerance test, and abdominal, thoracic, and thigh fat areas by CT. Total fat area (TFA) was calculated as the sum of these fat areas. Hypertension was defined as having a systolic blood pressure ≥140 mm Hg, having a diastolic blood pressure ≥90 mm Hg, or taking antihypertensive medications. Intra-abdominal fat area (IAFA) was associated with a higher prevalence of hypertension. Adjusted odds ratio of hypertension by IAFA was 1.68 for a 1-SD increase (95% CI, 1.20 to 2.37) after adjusting for age, sex, fasting plasma insulin, a nonlinear transformation of 2-hour plasma glucose, and TFA. IAFA remained a significant predictor of prevalence of hypertension even after adjustment for total subcutaneous fat area, abdominal subcutaneous fat area, body mass index, or waist circumference, but no measure of regional or total adiposity was associated with the odds of prevalence of hypertension in models that contained IAFA. Conclusions—Greater visceral adiposity increases the odds of hypertension in Japanese Americans independent of other adipose depots and fasting plasma insulin.

Diabetes and diabetes risk factors in second- and third-generation Japanese Americans in Seattle, Washington

In Seattle, Washington, the prevalence of diabetes was 20% in second-generation (Nisei) Japanese-American men and 16% in Nisei women 45-74 years old, while the prevalence of impaired glucose tolerance (IGT) was 36% in Nisei men and 40% in Nisei women. Hyperglycemia was less and duration of diabetes shorter in women. Related to diabetes and IGT in Nisei were higher fasting plasma insulin levels and central (visceral) adiposity. Prevalence of diabetes was low among the younger (34-53 years old) third-generation (Sansei) men and women. Among self-reported non-diabetic Sansei, however, prevalence of IGT was 19% in men and 29% in women, and IGT was associated with both increased fasting plasma insulin levels and more visceral fat, suggesting that many Sansei are at risk of future diabetes. An important lifestyle factor in the development of NIDD in Japanese Americans appeared to be dietary saturated (animal) fat. Another factor may be physical inactivity. In Japanese-American women, menopause also appeared to be an important risk factor. These risk factors may be related to fostering the accumulation of visceral fat and the development of insulin resistance. Five-year follow-up examinations performed in non-diabetic Nisei men and women have yielded additional information concerning the prognosis of IGT. Of those women who were IGT at baseline, 34% were diabetic at follow-up while 17% returned to normal. In men who had been IGT at baseline, 18% were diabetic at follow-up while 36% returned to normal. Over the 5-yr follow-up interval, proportionally more women progressed from normal to IGT (54%) then went from IGT to normal (17%). For men, roughly equal proportions went from normal to IGT (37%) as from IGT to normal (36%). It would therefore appear that greater proportions of Nisei women are progressing to IGT and to NIDD than are Nisei men. This observation may be related to the increased risk of developing central obesity and insulin resistance following menopause. Prevalence of cardiovascular disease (hypertension, peripheral vascular disease, and/or coronary heart disease) was increased in Japanese Americans with IGT and NIDD. Neuropathy and retinopathy were associated only with NIDD.

Low insulin secretion and high fasting insulin and C-peptide levels predict increased visceral adiposity : 5-year follow-up among initially nondiabetic Japanese-American men

Insulin resistance and hyperinsulinemia occur more frequently in subjects with greater visceral adiposity, but it is not known whether these metabolic abnormalities precede or follow visceral fat accumulation. We prospectively studied the development of visceral adiposity in relation to fasting and stimulated insulin and C-peptide levels. We followed 137 nondiabetic, second-generation Japanese-American men for changes in visceral adiposity over 5 years. Intra-abdominal fat (IAF) area (square centimeters) was measured at the umbilicus by computed tomography at baseline and after 5 years. Plasma insulin and C-peptide levels were measured after an overnight fast and during an oral glucose tolerance test, β-cell function was measured by the insulin secretion ratio (30–0 min plasma insulin difference)/(30–0 min plasma glucose difference). After adjustment for baseline IAF in multiple linear regression models, baseline fasting insulin (coefficient = 0.241, P = 0.048) and C-peptide (coefficient = 38.538, P < 0.001) levels were positively correlated, while the baseline insulin secretion ratio was negatively correlated with IAF change (coefficient = −0.099, P = 0.027). With IAF difference coded as a dichotomous variable (<0 cm2 vs. ≤0 cm2), the highest versus lowest tertile of baseline fasting insulin (odds ratio [OR] = 3.0, 95% CI 1.0–9.7) and fasting C-peptide (OR = 8.1, 95% CI 2.4–26.8) levels and the lowest versus highest tertile of the insulin secretion ratio (OR = 3.3, 95% CI 1.0–10.0) were associated with higher odds of IAF gain. Greater insulin resistance and reduced insulin secretion precede visceral fat accumulation in nondiabetic Japanese-American men.

Visceral Adiposity, Fasting Plasma Insulin, and Blood Pressure in Japanese-Americans

OBJECTIVE To examine the associations among blood pressure, body mass index (BMI), intra-abdominal fat, and fasting plasma insulin levels among nondiabetic subjects. RESEARCH DESIGN AND METHODS Second- (Nisei, n = 290) and third- (Sansei, n = 230) generation Japanese-American subjects without non-insulin-dependent diabetes mellitus (NIDDM) were selected from a community-based study of NIDDM incidence and complications. A cross-sectional comparison of measures obtained at the baseline visit was performed. Intra-abdominal fat (IAF) area was assessed using computed tomography. Associations among blood pressure, fasting insulin, and adiposity measures were assessed by comparison of mean values and multiple linear regression analysis. RESULTS— Hypertensive men and women had significantly higher mean IAF areas. Fasting insulin levels were somewhat higher in hypertensive subjects, with the only significant difference occurring among Sansei men. Both systolic and diastolic blood pressure correlated more strongly with IAF than BMI or skinfold thicknesses among Nisei, whereas among Sansei, IAF and BMI correlated equally well with either blood pressure. Significant positive correlations were found between fasting insulin level and blood pressure among Sansei only, even after adjustment for IAF and BMI (diastolic blood pressure - insulin coefficient = 0.24, P = 0.0043; systolic blood pressure insulin coefficient = 0.36, P = 0.0025). CONCLUSIONS IAF correlated more strongly with blood pressure than BMI or skinfold thicknesses among older, second-generation Japanese-Americans and was positively correlated with blood pressure among Sansei independent of fasting insulin level. Fasting insulin was significantly correlated with blood pressure independent of visceral and overall adiposity among third-generation Japanese-Americans.

Prenatal psychological stress, dermatoglyphic asymmetry and pregnancy outcome in the pigtailed macaque (Macaca nemestrina).

The relationships between maternal psychosocial stress during pregnancy and asymmetry in dermal ridge count, gestation length, birth weight percentile and survival were examined in 68 pigtailed macaque offspring. Twenty-five pregnant females were stressed daily by capture from 30 through 130 days postconception; 43 control unstressed females were housed under conditions of minimal disturbance. The difference between total intercore dermal ridge counts between right and left hands of the offspring was used as a measure of the perturbed development that theoretically occurs in the presence of a prenatal stressor. Dermatoglyphic asymmetry was significantly higher in the stressed offspring than in the unstressed group (mean asymmetry = 7.3 +/- 2.8 and 5.4 +/- 2.5, respectively; t = 2.85, p less than 0.01). Although maternal stress alone was not related to differential gestation length, birth weight, or survival, high asymmetry (8-13 residual dermal ridges) was significantly associated with increased perinatal mortality.

Association of plasma triglyceride and C-peptide with coronary heart disease in Japanese-American men with a high prevalence of glucose intolerance

SummaryIn a community-based study of second-generation Japanese-American men known to have a high prevalence of both Type 2 (non-insulin-dependent) diabetes and impaired glucose tolerance, there was a highly significant association of coronary heart disease with glucose intolerance in a study sample of 219 men. Intra-abdominal cross sectional fat area determined by computed tomography was significantly elevated in men with coronary heart disease even after adjustment for glucose intolerance and body mass index (p=0.026). Other differences that were significantly related to coronary heart disease after adjustment for glucose intolerance were lower high density lipoprotein cholesterol levels (p=0.001), elevated total triglyceride and very low density lipoprotein triglyceride (p<0.001), and elevated fasting insulin and C-peptide levels p=0.001. When these variables were tested in a stepwise multiple logistic regression model, significant independent associations with coronary heart disease were found only for total triglyceride and fasting C-peptide after adjustment for glucose tolerance status. Variables identified to be associated with coronary heart disease were interpreted as representing or manifesting an insulin resistant state. Thus, insulin resistance may be the underlying risk factor aetiologically linking glucose intolerance with coronary heart disease.