Hypertension and Type 2 Diabetes Comorbidity ... - Wiley Online Library

Hypertension and Type 2 Diabetes Comorbidity in Adults in the United States: Risk of Overall and Regional Adiposity Ike S. Okosun,* K. M. Dinesh Chandra,† Simon Choi,* Jacqueline Christman,*‡ G. E. Alan Dever,* and T. Elaine Prewitt§

Abstract OKOSUN, IKE S., K. M. DINESH CHANDRA, SIMON CHOI, JACQUELINE CHRISTMAN, G. E. ALAN DEVER, AND T. ELAINE PREWITT. Hypertension and type 2 diabetes comorbidity in adults in the United States: risk of overall and regional adiposity. Obes Res. 2001;9: 1–9. Objective: To evaluate the impact of generalized, abdominal, and truncal fat deposits on the risk of hypertension and/or diabetes and to determine whether ethnic differences in these fat patterns are independently associated with increased risk for the hypertension– diabetes comorbidity (HDC). Research Methods and Procedures: Data (n ⫽ 7075) from the Third U.S. National Health and Nutrition Examination Survey were used for this investigation. To assess risks of hypertension and/or diabetes that were due to different fat patterns, odds ratios of men and women with various cutpoints of adiposities were compared with normal subjects in logistic regression models, adjusting for age, smoking, and alcohol intake. To evaluate the contribution of ethnic differences in obesity to the risks of HDC, we compared blacks and Hispanics with whites. Results: Generalized and abdominal obesities were independently associated with increased risk of hypertension, diabetes and HDC in white, black, and Hispanic men and women. The risk of HDC due to generalized, truncal, and abdominal obesities tended to be higher in whites than

Submitted for publication April 3, 2000. Accepted for publication in final form August 21, 2000. Departments of *Community Medicine, †Internal Medicine, and ‡Family Medicine, Mercer University School of Medicine, Macon, Georgia; and §Department of Preventive Medicine and Epidemiology, Loyola University Medical Center, Stritch School of Medicine, Maywood, Illinois. Address correspondence to Ike S. Okosun, Department of Community Medicine, Mercer University School of Medicine, 1550 College Street, Macon, GA 31207. E-mail: [email protected] Copyright © 2000 NAASO

blacks and Hispanics. In men, the contribution of black and Hispanic ethnicities to the increased risk of HDC due to the various obesity phenotypes was ⬃73% and ⬃61%, respectively. The corresponding values for black and Hispanic women were ⬃115% and ⬃125%, respectively. Conclusions: In addition to advocating behavioral lifestyles to curb the epidemic of obesity among at-risk populations in the United States, there is also the need for primary health care practitioners to craft their advice to the degree and type of obesity in these at-risk groups. Key words: hypertension, diabetes, comorbidity

Introduction Hypertension and type 2 diabetes are interrelated metabolic disorders (1,2) that strongly predispose an individual to atherosclerotic cardiovascular disease (CVD) and to renal failure. (3,4) In the United States, an estimated 3 million individuals have both hypertension and diabetes (5), with hypertension being twice as common in subjects with diabetes compared to those without (2,5). Indeed, an estimated 35% to 75% of subjects with diabetic CVD and renal disease are known to have hypertension (1,2). The prevalence of coexisting hypertension and diabetes is on the rise in the United States, and may be due to a comparable increase in obesity. Although different anthropometric indices of obesity are generally connected with hypertension and diabetes (6 –9), the mechanisms and independent roles of these indices are not clear. The most commonly used anthropometric index of obesity is the body mass index (BMI). BMI is a heightadjusted measure of overall body heaviness that is highly correlated with adiposities measured by computerized tomography, dual-energy X-ray absorptiometry, and doubly labeled water (10). However, BMI has been found to lack sufficient explanatory power for many cardiovascular disorders (11–13). It is now being recognized that other aberOBESITY RESEARCH Vol. 9 No. 1 January 2001

1

Coexistence of Hypertension and Diabetes, Okosun et al.

rant regional fat distributions as seen in abdominally or centrally obese subjects may be more potent in CVD than overall heaviness as determined by BMI (14,15). The regional body habitus most commonly linked to obesities are waist girth and skinfolds, such as subscapular and triceps. Waist-to-hip ratio (WHR) is also sometimes used as an anthropometric measure of abdominal fat accumulation (16). However, WHR is not universally useful as a surrogate for predicting the visceral amounts of adipose tissue that are most metabolically responsible for the obesity-related disease risk or incidence of disease (17). Only limited data exist regarding the association between generalized and regional fat deposits and an increased risk of hypertension– diabetes comorbidity (HDC). Because of insufficient sample size, many studies presented to date are restricted to either hypertension or diabetes, and not both. Also, studies comparing risks of HDC due to regional fat deposits in ethnic groups in the United States are few despite contrasting degrees of susceptibility to hypertension and diabetes. In this study, we aimed to evaluate the risks of HDC due to generalized and regional adiposities. We also sought to determine whether or not ethnic differences in obesity phenotypes were independently associated with the risk of HDC.

Research Methods and Procedures Data Source Data from the Third U.S. National Health and Nutrition Examination Survey (NHANES III) as provided by the National Center for Health Statistics were used in this investigation. The sampling and measurement procedures have been described in detail previously (18 –20). Briefly, NHANES III is a complex, multistage probability sample of noninstitutionalized U.S. population groups defined and examined between 1988 and 1994. Only subjects identified as non-Hispanic white, non-Hispanic black, and Hispanic Americans were selected for this investigation. This study was further confined to 7075 individuals aged 40 to 75 years for whom anthropometric variables, including weight, height, waist, hip, and diastolic blood pressure (DBP) and systolic blood pressure (SBP) were obtained, and for whom fasting plasma blood glucose and skinfold thicknesses had been measured. Weight was measured at a standing position using a Toledo self-zeroing weight scale (Seritex, Carlstadt, NJ). Height was measured at an upright position with a stadiometer. Waist measurement was made at the natural waist midpoint between the lowest aspect of the rib cage and highest point of the iliac crest, and to the nearest 0.1 cm. Three blood pressure measurements were obtained from each subject using a standard mercury sphygmomanometer at 60-second intervals between inflation cuffs. The average of the three readings was used for this analysis. Smoking 2

OBESITY RESEARCH Vol. 9 No. 1 January 2001

and alcohol intake were assessed by self-reporting. Smoking was categorized as 1 and 0 for current smokers and nonsmokers, respectively. Alcohol use was graded as 1 for current drinkers and 0 for current nondrinkers. Details of the methods that were used for blood collection and plasma glucose assays have been described elsewhere by other investigators (20,21). Briefly, plasma glucose was measured using a modified hexokinase enzymatic method (20,21). Skinfolds Measurement Skinfold thicknesses were measured with Holtain skinfold calipers (Holtain, Crymych, UK) at the right side of the body with subjects in upright position (22). Folds of the skin and the underlying subcutaneous adipose tissue were gently grasped between the left thumb and forefingers 2.0 cm above the place measurement was taken. The jaws of the calipers were placed perpendicular to the length of the fold, and the skin thickness was measured to the nearest 0.1 mm while the fingers continued to hold the skinfold. Actual measurements were read from the caliper ⬃3 seconds after the caliper tension was released (22). Subscapular Skinfold. After palpating the anterior angle of the right scapular, a mark was made at the interior angle of the scapular with a cosmetic pencil marker. A fold of skin directly 1.0 cm below and medial to the interior angle ⬃45 degrees below the horizontal extending toward the right elbow was used (22). Triceps Skinfold. With the subject’s shoulders relaxed and arms hanging freely at the sides, triceps skinfold was measured at the point on the posterior surface of the upper arm (22). Definition of Terms Based on American Diabetes Association criteria (23), type 2 diabetes is defined as either current diagnosis and use of insulin or hypoglycemic agent or fasting blood glucose of ⱖ126 mg/dL or a 2-hour postload in the oral glucose tolerance test of ⬎200 mg/dL. Hypertension was defined as SBP of ⱖ140 or DBP of ⱖ90 or current use of antihypertension medication (24). HDC was defined as the existence of both hypertension and diabetes. Abdominal obesity was defined as waist circumference of ⱖ102 cm and ⱖ88 cm for men and women, respectively (25,26). Overweight was defined as a BMI between 25 and 29.99 kg/m2 and obesity was defined as a BMI of ⱖ30 kg/m2 for both men and women (25). An index of fat deposition patterns (truncal obesity) was calculated from skinfolds. Truncal obesity is the ratio of subscapular to triceps skinfold (STR) (27). Cut-Points for Truncal Adiposity To determine the cut-points for truncal obesity, we used gender-specific empirical quartile distributions of STR. Mean values for the highest quartiles were employed as


appropriate cut-points for these fat deposition patterns. The estimated cut-points of STR for truncal obesity were ⱖ2.50 mm and ⱖ1.34 mm for men and women, respectively. Statistical Analysis Statistical programs available in SPSS for Windows were used for this analysis (28). Ethnic differences for continuous and categorical variables were assessed by one-way ANOVA and ␹2 statistics, respectively. Tukey’s range test and ␹2 test were used to compare means and prevalences between ethnic groups, respectively. Prevalence estimates were weighted to account for cluster design and to represent the total civilian noninstitutionalized population of the United States. The prevalence of hypertension and diabetes, HDC, and generalized and regional obesities were ageadjusted by direct methods using the 1990 U.S. population census data. To assess the risk of HDC that was associated with generalized and regional fat patterns, we compared odds ratios of men and women in the various obesity cut-points with normal weight subjects in logistic regression models, adjusting for age, smoking, and alcohol intake. To determine the contribution of ethnic differences in generalized and regional adiposities to the risk of HDC, dummy variables were used to compare non-whites with whites fitted in gender-specific logistic regression models, adjusting for age, smoking, and alcohol intake.

Results A total number of 3261 whites, 1924 blacks, and 1890 Hispanic men and women were eligible for this investigation. Gender-specific mean values of age, anthropometric variables, skinfold thicknesses, blood pressures, and plasma glucose, as well as prevalences of obesity phenotypes, hypertension, and diabetes are shown in Tables 1 and 2 for men and women, respectively. Overall, there were ethnic differences for all variables, with white men and women tending to be older than their Black and Hispanic counterparts (p ⬍ 0.001). White men were heavier (in terms of weight and BMI) and presented with larger triceps skinfold thickness and lower STR than black and Hispanic men (p ⬍ 0.001). Black men presented with lower waist girth and higher mean values of DBP and SBP compared with white and Hispanic men (p ⬍ 0.001). The mean values of WHR and fasting and postprandial glucose were higher in Hispanic men compared with white and black men (p ⬍ 0.001). Black women were heavier as determined by weight and BMI, and presented with larger waist girth; subscapular, triceps, and STR skinfold thicknesses; DBP; and SBP than white and Hispanic women (p ⬍ 0.001). Hispanic women presented with higher mean values of WHR and postprandial glucose values compared with their white and black counterparts (p ⬍ 0.001).

Hispanic men presented with higher prevalences of generalized obesity, diabetes, and HDC compared with whites and blacks (p ⬍ 0.01). The prevalences of abdominal obesity and hypertension were higher among white and black men, respectively. Fewer white women were estimated to have generalized, abdominal, and truncal obesities compared with their black and Hispanic counterparts. Whereas higher prevalences of generalized obesity and hypertension were noted among black women, Hispanic women presented with the greatest proportions of overweight, diabetic, and HDC subjects. We fitted gender- and sex-specific logistic regression models separately for hypertension and diabetes, adjusting for age, smoking, and alcohol intake (Table 3). In men and women, generalized and abdominal obesities were independently associated with increased risk of hypertension and diabetes in the three ethnic groups. In men, risks of hypertension due to generalized and abdominal obesities were higher in blacks, whereas in women the risk tended to be higher in whites. The odds ratios for hypertension due to adiposity phenotypes ranged from ⬃1 for truncal obesity in Hispanic men to ⬃2.2 for generalized obesity in black men. The corresponding values for diabetes ranged from ⬃1 for truncal obesity in Hispanic men to ⬃2.6 for generalized obesity in black men. In women, odds ratios for hypertension ranged from 1 for truncal obesity in Hispanics to 3.7 for generalized obesity in whites. The analogous values for diabetes in women were ⬃1 and ⬃2.3 for truncal and generalized obesities, respectively. Results of ethnic-specific multiple logistic regression models showing odds ratios for HDC due to generalized and regional obesities are presented in Table 4. In the three ethnic groups, the risk of HDC due to obesity phenotypes tended to be greater in women compared with men. Also, the risk associated with HDC tended to be higher in white men and women compared with their black and Hispanic counterparts. The risk for HDC due to generalized obesity ranged from 1.7 in Hispanic men to 3.2 in white men, and from 1.9 in black women to 3.0 in white women. The analogous values for abdominal obesity ranged from 1.8 in Hispanic men to 2.6 in white men and from 2.5 in Hispanic women to 4.0 in white women. The risk for HDC due to truncal obesity ranged from ⬃1 in Hispanic men to ⬃1.6 in white men and from 1.1 in Hispanic women to 1.6 in white women. To determine whether ethnic differences in the various obesity phenotypes or fat patterns were associated with risks of HDC independent of age, smoking, and alcohol consumption, dummy variables were used to compare blacks and Hispanics with whites in sex-specific logistic linear regression models (Table 5). For both men and women, black and Hispanic ethnicities were independently associated with greater risk of HDC relative to white ethnicity. The risk for HDC due to black and Hispanic ethnicities was OBESITY RESEARCH Vol. 9 No. 1 January 2001

3


Table 1. Characteristics of studied variables in men Variables n (%) Age (year) Weight (kg) BMI (kg/m2) Waist circumference (cm) WHR Skinfold thicknesses (mm) Subscapular Triceps STR

Whites

Blacks

Hispanics

p value

1574 (45.7) 56.6 ⫾ 10.4b 84.8 ⫾ 15.8a 27.4 ⫾ 4.7a 100.7 ⫾ 11.9a 1.00 ⫾ 0.06a

906 (26.3) 55.7 ⫾ 10.4a 82.4 ⫾ 17.4b 26.8 ⫾ 5.0b 96.1 ⫾ 13.7b 0.97 ⫾ 0.06b

962 (27.9) 55.7 ⫾ 10.3a 80.6 ⫾ 14.6c 28.2 ⫾ 4.5c 100.4 ⫾ 11.4a 1.01 ⫾ 0.06c

⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001

20.6 ⫾ 7.3a 14.0 ⫾ 6.1b 1.61 ⫾ 0.53b

20.4 ⫾ 8.6a 12.3 ⫾ 6.7a 1.86 ⫾ 0.58a

21.5 ⫾ 7.0b 12.8 ⫾ 5.5a 1.88 ⫾ 0.56a

0.003 ⬍0.001 ⬍0.001

Blood pressure (mm Hg) DBP SBP

76.8 ⫾ 10.7a 128.7 ⫾ 17.8a

80.1 ⫾ 12.4b 134.0 ⫾ 20.9b

77.6 ⫾ 10.9a 130.3 ⫾ 19.9a

⬍0.001 ⬍0.001

Plasma glucose (mg/dL) Fasting glucose Postprandial

106.2 ⫾ 35.4a 143.4 ⫾ 71.3a

109.7 ⫾ 49.9a 139.0 ⫾ 75.2a

115.2 ⫾ 50.4b 164.2 ⫾ 93.2b

⬍0.001 ⬍0.001

Prevalences (%) Overweight* Generalized obesity† Abdominal obesity‡ Truncal obesity§ Diabetes# Hypertension㛳 HDC

44.5a 24.5a 42.6a 15.4b 11.1a 46.1a 12.5a

40.6b 22.1a 28.2b 28.9a 13.1b 57.7b 16.4b

44.9a 31.3c 40.1c 29.3a 16.3c 43.8c 17.0c

⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 0.003

Values are means ⫾ SDs for continuous variables. * Overweight was defined as a BMI of between 25 and 29.99 kg/m2. † Generalized obesity was defined as a BMI of ⱖ30 kg/m2. ‡ Abdominal obesity was defined as a waist circumference of ⱖ102 cm. § Truncal obesity was derived from the STR; mean value (ⱖ2.50 mm) for the highest quartile was employed for truncal obesity cut-point. # Diabetes was defined as either a) current diagnosis and use of insulin or hypoglycemic agent or b) a fasting plasma glucose value of ⱖ126 mg/dL or a 2-hour postload in the oral glucose tolerance test of ⬎200 mg/dL. 㛳 Hypertension was defined as SBP of ⱖ140 or DBP of ⱖ90 or current use of antihypertension medication; overall p value is from one-way ANOVA across groups; values with different superscripts differ significantly (p ⬍ 0.05) in pair-wise comparisons.

⬃1.7 for men and ⬃2.0 for women, respectively. Increase in age was also associated with enhanced risk of HDC, and the odds ratios tended to stabilize at 1.05 for men and 1.08 for women. Finally, to determine whether overweight as defined by a BMI of between 25 and 29.99 was associated with increased risk of hypertension, diabetes, or HDC, we also fitted ethnic- and gender-specific logistic regression models adjusting for age, smoking, and alcohol intake (data not shown). 4


The results indicate that being overweight was not significantly associated with increased risk for hypertension, diabetes, or HDC.

Discussion Hypertension, type 2 diabetes, and obesity are clinical features with a multifactoral and complex pathophysiology (29 –33). Although the relationship of obesity with hyper-


Table 2. Characteristics of studied variables in women Variables

Whites

Blacks

Hispanics

p value

n (%) Age (year) Weight (kg) BMI (kg/m2) Waist circumference (cm) WHR

1687 (46.4) 57.6 ⫾ 10.3a 71.5 ⫾ 16.3a 27.4 ⫾ 6.1a 93.1 ⫾ 14.6a 0.89 ⫾ 0.08a

1018 (28.0) 54.1 ⫾ 10.1b 80.3 ⫾ 19.9b 30.3 ⫾ 7.2b 98.4 ⫾ 15.5b 0.91 ⫾ 0.08b

928 (25.5) 55.9 ⫾ 10.3c 71.0 ⫾ 14.6a 29.3 ⫾ 5.8c 96.8 ⫾ 12.7c 0.93 ⫾ 0.08c

⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001

Skinfold thicknesses (mm) Subscapular Triceps STR

22.7 ⫾ 8.9a 25.2 ⫾ 7.8a 0.91 ⫾ 0.27a

27.3 ⫾ 9.4b 27.0 ⫾ 9.0b 1.07 ⫾ 0.29b

25.6 ⫾ 7.8c 25.8 ⫾ 7.3a 1.02 ⫾ 0.26c

⬍0.001 ⬍0.001 ⬍0.001

Blood pressure (mm Hg) DBP SBP

71.3 ⫾ 10.7a 125.2 ⫾ 19.8a

74.6 ⫾ 13.2b 132.3 ⫾ 23.6b

72.0 ⫾ 11.0a 129.0 ⫾ 22.1c

⬍0.001 ⬍0.001

Plasma glucose (mg/dL) Fasting glucose Postprandial

102.9 ⫾ 38.9a 147.4 ⫾ 70.5a

111.9 ⫾ 55.2b 148.8 ⫾ 83.7a

117.8 ⫾ 56.7c 178.8 ⫾ 94.5b

⬍0.001 ⬍0.001

Prevalences (%) Overweight* Generalized obesity† Abdominal obesity‡ Truncal obesity§ Diabetes# Hypertension㛳 HDC

30.9a 28.4a 57.9a 15.5b 10.0a 43.0a 13.0a

32.1a 46.3b 70.5b 27.2a 17.6b 61.3b 20.6b

38.4b 39.2c 71.2b 26.2a 22.1c 46.3c 22.4c

⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001

Values are means ⫾ SDs for continuous variables. * Overweight was defined as a BMI of between 25 and 29.99 kg/m2. † Generalized obesity was defined as a BMI of ⱖ30 kg/m2. ‡ Abdominal obesity was defined as a waist circumference of ⱖ88 cm. § Truncal obesity was derived from the STR; mean value (ⱖ1.34 mm) for the highest quartile was employed for truncal obesity cut-point. # Diabetes was defined as either a) current diagnosis and use of insulin or hypoglycemic agent or b) fasting plasma glucose value of ⱖ126 mg/dL or a 2-hour postload in the oral glucose tolerance test of ⬎200 mg/dL. 㛳 Hypertension was defined as SBP of ⱖ140 or DBP of ⱖ90 or current use of antihypertension medication; overall p value is from one-way ANOVA across groups; values with different superscripts differ significantly (p ⬍ 0.05) in pair-wise comparisons.

tension and diabetes is not completely clear, it is well known that excess fat leads to increased insulin resistance (29) and that insulin resistance predisposes to diabetes (29 – 33). Most of these assertions about the role of obesity are based on overall body fatness. However, it is clear that various regional fat distributions may have different metabolic implications. While most of these observations are based on independent effects of these obesity phenotypes or

fat patterns on hypertension and diabetes (34 –37), little information exists regarding the roles of these various regional fat deposits on the joint existence of hypertension and diabetes. To our knowledge, this is the first investigation using NHANES III to determine the prevalence of HDC in the three major ethnic groups in the United States. NHANES III represents the best available data because the sampling OBESITY RESEARCH Vol. 9 No. 1 January 2001

5


Table 3. Ethnic-specific odds ratios for hypertension and diabetes due to generalized and regional obesity White Hypertension Obesity phenotype OR

95% CI

Black Diabetes

OR

95% CI

Hypertension OR

95% CI

Hispanic Diabetes

OR

95% CI

Hypertension OR

Diabetes

95% CI

OR

95% CI

Men Generalized obesity 2.21 1.68–2.91 2.39 1.80–3.16 2.24 1.52–3.24 2.60 1.77–3.81 2.10 1.54–2.86 1.73 1.33–2.52 Abdominal obesity 1.97 1.59–2.44 2.40 1.84–3.14 2.14 1.55–2.96 2.24 1.57–3.20 1.94 1.47–2.56 1.86 1.31–3.28 Truncal obesity 1.18 0.89–1.57 1.44 1.02–2.01 1.03 0.76–1.39 1.06 0.74–1.53 0.80 0.60–1.08 0.99 0.72–1.38 Women Generalized obesity 3.67 2.88–4.67 2.31 1.77–3.03 1.41 1.14–1.99 2.04 1.49–2.78 1.94 1.44–2.63 1.72 1.28–2.32 Abdominal obesity 3.25 2.59–4.06 2.22 1.05–3.08 1.50 1.04–1.91 2.38 1.61–3.51 2.04 1.47–2.83 1.74 1.25–2.43 Truncal obesity 1.50 1.13–1.97 1.45 1.08–2.00 1.30 1.03–1.93 1.48 1.04–2.01 1.00 0.72–1.37 1.04 1.02–1.45 OR, odds ratio obtained from age-, smoking-, and alcohol-adjusted logistic regression model; CI, confidence intervals.

In this study, abdominal obesity was the most powerful obesity phenotype associated with HDC risk in women. Black and Hispanic ethnicities were independently associated with greater risk of HDC relative to whites. The contribution of men of black and Hispanic ethnicities to the increased risk of HDC due to the various obesity phenotypes was ⬃73% and ⬃61%, respectively. The corresponding values for black and Hispanic women were ⬃115% and ⬃125%, respectively. The observed risks for hypertension and diabetes due to the obesity phenotypes that were noted for this investigation are consistent with previous findings (35,36). Similar to the results of this investigation, Ko et al. (38) also noted a

scheme was representative and national in scope. The training program and quality control measures instituted in NHANES III give an added credibility to the data. The results of this investigation show that generalized and abdominal obesities are associated with increased risks of HDC in the three ethnic groups. With the exception of Hispanic men, truncal obesity was associated with increased risk for HDC in the three ethnic groups. Generalized and abdominal obesities were also independently associated with increased risk for hypertension and diabetes in black, white, and Hispanic men and women. In whites and blacks, truncal obesity was associated with increased risk of hypertension, adjusting for age, smoking, and alcohol intake.

Table 4. Ethnic-specific odds ratios for HDC due to generalized and regional obesity White Obesity phenotype

OR

Black

95% CI

Hispanic

OR

95% CI

OR

95% CI

1.97 2.41 1.11

1.59–3.62 1.63–3.54 1.04–1.06

1.74 1.81 0.75

1.19–2.54 1.06–2.17 0.50–1.12

1.40–2.77 1.30–2.98 1.11–2.26

2.00 2.51 1.13

1.42–2.82 1.62–3.89 0.78–1.63

Men Generalized obesity Abdominal obesity Truncal obesity

3.20 2.63 1.60

2.31–4.40 1.91–3.63 1.08–2.35

Women Generalized obesity Abdominal obesity Truncal obesity

2.99 3.99 1.64

2.32–4.39 2.07–4.34 1.13–2.39

1.97 2.97 1.58

OR, odds ratio obtained from age-, smoking-, and alcohol-adjusted logistic regression model; CI, confidence intervals.

6



Table 5. Effect of ethnicity on the risk of generalized and regional obesity and HDC Adiposity OR

95% CI

Age OR

95% CI

Smoking OR

95% CI

Alcohol OR

95% CI

Black OR

95% CI

Hispanic OR

95% CI

Men Generalized obesity 2.43 1.97–2.99 1.06 1.04–1.07 0.81 0.64–1.02 0.76 0.62–0.93 1.71 1.33–2.09 1.61 1.27–2.04 Abdominal obesity 2.13 1.74–2.61 1.05 1.04–1.06 0.78 0.62–0.99 0.77 0.63–0.94 1.86 1.45–2.39 1.61 1.27–2.04 Truncal obesity 1.09 0.86–1.38 1.05 1.04–1.06 0.72 0.57–0.90 0.76 0.62–0.92 1.65 1.29–2.11 1.65 1.30–2.09 Women Generalized obesity 2.37 1.96–2.88 1.09 1.07–1.10 0.94 0.74–1.19 0.71 0.56–0.90 2.05 1.62–2.59 2.09 1.73–2.76 Abdominal obesity 2.53 2.00–3.20 1.08 1.07–1.09 0.89 0.70–1.14 0.71 0.56–0.89 2.17 1.72–2.74 2.13 1.69–2.69 Truncal obesity 1.41 1.43–1.75 1.08 1.07–1.09 0.83 0.65–1.06 0.67 0.53–0.85 2.25 1.78–2.83 2.19 1.74–2.76 OR, odds ratio obtained from logistic regression model; CI, confidence intervals.

significant association of abdominal and generalized obesities with hypertension in Chinese subjects. Our findings are consistent with the Framingham and Coronary Artery Risk Development in Young Adults studies, both of which found that generalized and regional obesities promote CVD risks (39,40). As in the Coronary Artery Risk Development in Young Adults study, our study shows a higher prevalence of obesity in black women. Also, black women presented with greater mean levels of STRs and larger waist girths than white women (41). The lower risk for hypertension and HDC due to generalized, abdominal, and truncal obesities in blacks corroborates the findings of other investigators (15,42,43) in which obesity was found to be less strongly associated with cardiovascular risks in blacks than whites. Similar to the San Antonio Heart Study (44), our investigation indicates that abdominal obesity was the strongest predictor of diabetes and HDC among Hispanic men and women compared with generalized or truncal obesities. Our findings of a lower risk of HDC in the Hispanic group due to truncal obesity are similar to the findings of Marshall et al. (45) Analysis of ethnic differences in risk factors associated with the prevalence of type 2 diabetes in the San Luis Valley Diabetes Study found that a unit increase in STR was associated with lower risk for type 2 diabetes among Hispanic compared with non-Hispanic whites (45). The stronger risk for HDC due to abdominal obesity compared with truncal obesity as noted in this study may be attributed to the very strong correlation between abdominal obesity and visceral obesity. This stronger risk for HDC due to abdominal obesity is consistent with the physiological and endocrinological bases of visceral adiposity (46,47). Visceral adiposity is more sensitive to lipolytic stimuli, and thus may have a more diminished antilipolytic effect on HDC than truncal obesity (46). It has been shown that visceral fat accumulation is strongly regulated by endocrine

mechanisms. Due to higher blood flow and abundant innervations, hormonal effects on adipose tissues are more pronounced in visceral than other fat depots. Thus, any systemic alteration in available lipid substrate or hormone secretion has more pronounced consequences on the visceral fat microenvironment (47). Also, the density of hormone receptors is higher in visceral fat than in other depots (47). As proposed by some investigators, the mechanism linking obesity with metabolic syndrome and the subsequent development of CVD may involve two processes. One, through the preponderance of the highly sensitive lipolytic hypertrophied visceral adipose tissues, elevated free fatty acids are suggested to induce insulin resistance (48,49). Two, elevated free testosterone and reduced sex hormonebinding globulin may promote increased abdominal adiposity and reduce fractional hepatic extraction of insulin (50). Thus, the connection of adiposity with HDC may be a result of the enlarged visceral fat depot discharging free fatty acids into the portal and systemic circulation (50). The higher odds ratios for HDC in whites, particularly white women, in this study are supported by findings by Dowling et al. (51) who examined effects of body fat distribution and race on insulin resistance in adipocytes of obese women. Our study correlates well with their findings that upper body obesity in white women but not in black women is more associated with insulin resistance (51). In the study by Dowling et al. (51), abdominal fat cells from the upper body obesity group vs. the lower body obesity group were less responsive to the stimulatory effects of insulin on glucose uptake and less sensitive to the antilipolytic effects of insulin in white women. In contrast, fat cells from upper and lower body groups were equally sensitive to the stimulatory effects of insulin on glucose transport and the suppressive effects of insulin on lipolysis. Thus, resisOBESITY RESEARCH Vol. 9 No. 1 January 2001

7


tance to the antilipolytic effects of insulin at the level of adipose tissue may increase systemic lipolysis and play a much higher role in HDC in abdominally obese white women than in abdominally obese black women. Some limitations relevant to the interpretation of the results from this study are noteworthy. First, although bias due to survey nonresponse and missing values for some variables cannot be ruled out, previous studies of NHANES have shown little bias due to nonresponse (52). Second, as a cross-sectional study, the directionality of the associations between HDC and adiposity cannot be clearly established. However, many epidemiological studies show obesity to precede the onset of either hypertension or diabetes (29 –33). Third, several wellknown hypertension and diabetes risk factors, including physical activity, were not examined. Epidemiological studies show an inverse relationship between physical activity or fitness and the incidence of hypertension and diabetes (53,54). Thus, unmeasured risk factors may have significant implications in the observed risks and the magnitude of the observed differences of the association between obesity phenotypes and hypertension, diabetes, or HDC in the different ethnic groups. Fourth, ethnic differences in the prevalence of hypertension in this study, despite the fact that the mean values of blood pressures were within acceptable ranges, were due to differences in the proportion of antihypertension medication use in these populations. Fifth, many of the obesity phenotypes that were investigated in this study carry the components of overall obesity. Because of the high colinearity of these variables, we did not adjust for the component that is carried by BMI.

Conclusions Hypertension and diabetes are complicated metabolic disorders. Separately, each condition is a potent risk factor for CVD, and together they strongly predispose to numerous sequalae, including end-stage renal disease, coronary artery disease, peripheral vascular disease, and CVD (1–5). Although none of the ethnic groups in this study may be classified as hypertensive or abdominally obese on the basis of the cut-points established for abdominal obesity, the proportion of subjects with hypertension and abdominal obesity in each group is clinically significant and warrants interventions. Because the results of this investigation show that the risk of HDC did not increase continuously with BMI in overweight subjects, there is a need for primary health care practitioners to craft their advice to the degree and type of obesity rather to degree of overweight. Because the prevalence of obesity is increasing dramatically in the United States, it is conceivable that the prevalence of these comorbidities and the associated sequalae will increase pari passu. Because the risk of HDC is closely associated with obesity, an important public health challenge is defining behavioral lifestyles to alleviate the epidemic of obesity in the United States. Weight management programs targeted to high-risk populations are needed for such 8


weight-related illnesses as diabetes and hypertension. Those strategies for weight reduction and maintenance should become a public health priority.

Acknowledgments Data from the NHANES III were obtained from the U.S. National Center for Health Statistics. References 1. Bakris G, Sowers J, Epstein M, Williams M. Hypertension in patients with diabetes. Why is aggressive treatment essential? Postgrad Med. 2000;107:53– 6, 61– 4. 2. Sowers JR, Epstein M. Diabetes mellitus and associated hypertension, vascular disease, and nephropathy. An update. Hypertension. 1995;26(6 Pt 1):869 –79. 3. Epstein M, Sowers JR. Diabetes mellitus and hypertension. Hypertension. 1992;19:403–18. 4. Weir MR. Diabetes and hypertension: blood pressure control and consequences. Am J Hypertens. 1999;12(12 Pt 1–2): 170S– 8S. 5. The National High Blood Pressure Education Program Working Group. The National High Blood Pressure Education Program Working Group report on hypertension in diabetes. Hypertension. 1994;23:145–58. 6. Fiori G, Facchini F, Pettener D, Rimondi A, Battistini N, Bedogni G. Relationships between blood pressure, anthropometric characteristics, and blood lipids in high- and low-altitude populations from Central Asia. Ann Hum Biol. 2000;27:19 –28. 7. Resnick HE, Valsania P, Halter JB, Lin X. Differential effects of BMI on diabetes risk among black and white Americans. Diabetes Care. 1998;21:1828 –35. 8. Rotimi CN, Cooper RS, Okosun IS, et al. Prevalence of diabetes and impaired glucose tolerance in Nigerians, Jamaicans, and U.S. blacks. Ethn Dis. 1999;9:190 –200. 9. Okosun IS, Cooper RS, Rotimi CN, Osotimehin B, Forrester T. Association of waist circumference with risk of hypertension and type 2 diabetes in Nigerians, Jamaicans, and African-Americans. Diabetes Care. 1998;21:1836 – 42. 10. Despres JP, Ross R, Lemeieux S. Imaging techniques applied to the measurement of human body composition. In: Roche AF, Heymsfield SB, Lohman TG, eds. Human Body Composition. Champaign, IL: Human Kinetics; 1996, pp. 149 – 66. 11. Garrow J. Body composition for the investigation of obesity. Basic Life Sci. 1990;55:183–90. 12. Cerovska J, Petrasek R, Hajnis K, Kaucka J. Indices of obesity and body composition in four groups of Czech population. Z Morphol Anthropol. 1977;68:213–9. 13. Vanhala MJ, Kumpusalo EA, Pitkajarvi TK, Takala JK. Metabolic syndrome in a middle-aged Finnish population. J Cardiovasc Risk. 1997;4:291–5. 14. Banerji MA, Chaiken RL, Gordon D, Kral JG, Lobvitz HE. Does intra-abdominal adipose tissue in black men determine whether NIDDM is insulin-sensitive? Diabetes. 1995; 44:141– 6. 15. Conway JM, Yanovski SZ, Avila NA, Hubbard VS. Visceral adipose tissue differences in black and white women. Am J Clin Nutr. 1995;61:765–71.


16. Molarius A, Siedell JC. Selection of anthropometric indicators for classification of abdominal fatness: a critical review. Int J Obes Relat Metab Disord. 1998;22:719 –27. 17. Lemieux S, Prud’homme D, Bouchard C, Tremblay A, Despres JP. A single threshold value of waist girth identifies normal-weight and overweight subjects with excess visceral adipose tissue. Am J Clin Nutr. 1996;64:685–93. 18. Ezzati TM, Massey JT, Waksberg J, Chu A, Maurer KR. Sample design: Third National Health and Nutrition Examination Survey. Vital Health Stat 2. 1992;113:1–35. 19. National Center for Health Statistics. National Health and Nutrition Examination Survey III: Data Collection Forms. Hyattsville, MD: National Center for Health Statistics; 1991. 20. Johnson CL, Rifkind BM, Sempos CT, et al. Declining serum total cholesterol levels among US adults. The National Health and Nutrition Examination Surveys. JAMA. 1993;269:3002– 8. 21. Harris MI, Flegal KM, Cowie CC, et al. Prevalence of diabetes, impaired fasting glucose, and impaired glucose tolerance in U.S. adults. The Third National Health and Nutrition Examination Survey, 1988 –1994. Diabetes Care. 1998:21:518 –24. 22. Westat Inc. National Health and Nutrition Examination Survey III. Body Measurements (Anthropometry). Rockville, MD: National Center for Health Statistics; 1996. 23. American Diabetes Association. Screening for Type 2 diabetes. Diabetes Care. 1998;12(Suppl 1):S20 –2. 24. 1988 Joint National Committee. The 1988 Report of the Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure. Arch Intern Med. 1988;148:1023–38. 25. National Institutes of Health/National Heart, Lung, and Blood Institute. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults. The evidence report. Obes Res. 1998;6:51S–209S. 26. Lean ME, Han TS, Morrison CE. Waist circumference as a measure for indicating need for weight management. BMJ. 1995;311:158 – 61. 27. Ballard-Barbash R, Schatzkin A, Carter CL, et al. Body fat distribution and breast cancer in the Framingham Study. J Natl Cancer Inst. 1990;82:286 –90. 28. Norusis MJ. SPSS for Windows Release 8.0. Chicago: SPSS Inc.; 1998. 29. Kolterman OG, Insel J, Seakow M, Olefsky JM. Mechanisms of insulin resistance in human obesity: evidence for receptor and postreceptor defects. J Clin Invest. 1980;65:1272– 84. 30. Despres J-P. Dyslipidaemia and obesity. Baillieres Clin Endocrinol Metab. 1994;8:629 – 60. 31. Haffner SM, Valdez RA, Hazed HP, Mitchell BD, Morales PA, Stern MP. Prospective analysis of the insulin-resistance syndrome (syndrome X). Diabetes. 1992;41:715–22. 32. Hanefeld M, Leonhardt W. Das metabolische syndrom. Dtsch Gesundh Wesen. 1981;36:545–51. 33. Reaven GM. Banting Lecture 1988. Role of insulin resistance in human disease, 1988. Nutrition. 1997;13:65. 34. Weiderpass E, Persson I, Adami HO, Magnusson C, Lindgren A, Baron JA. Body size in different periods of life, diabetes mellitus, hypertension, and risk of postmenopausal endometrial cancer. Cancer Causes Control. 2000;11:185–92. 35. Khaodhiar L, McCowen KC, Blackburn GL. Obesity and its comorbid conditions. Clin Cornerstone. 1999;2:17–31.

36. Sowers JR. Obesity and cardiovascular disease. Clin Chem. 1998;44(8 Pt 2):1821–5. 37. Han TS, Feskens EJ, Lean ME, Seidell JC. Associations of body composition with type 2 diabetes mellitus. Diabet Med. 1998;15:129 –35. 38. Ko GT, Chan JC, Woo J, et al. Simple anthropometric indexes and cardiovascular risk factors in Chinese. Int J Obes Relat Metab Disord. 1997;21:995–1001. 39. Kannel WB, Cupples LA, Ramaswani R, Stroke J, Kreger BE, Higgins M. Regional obesity and risk of cardiovascular disease: the Framingham study. J Clin Epidemiol. 1991;44: 183–90. 40. Burke GL, Jacobs DR, Sprafka JM, Savage PJ, Sidney S, Wagenknecht LE. Obesity and overweight in young adults: the CARDIA study. Prev Med. 1990;19:476 – 88. 41. Folsom AR, Burke GL, Byers CL, et al. Implication of obesity for cardiovascular disease in blacks: the CARDIA and ARIC studies. Am J Clin Nutr. 1991;53:1604S–11S. 42. Lovejoy JC, de la Bretonne JA, Klemperer M, Tulley R. Abdominal fat distribution and metabolic risk factors: effects of race. Metabolism. 1996;45:1119 –24. 43. Albu J, Murphy L, Frager DH, Johnson JA, Pi-Sunyer FX. Visceral fat and race-dependent health risks in obese nondiabetic premenopausal women. Diabetes. 1997;46:456 – 62. 44. Haffner SM. Obesity and the metabolic syndrome: the San Antonio Heart Study. Br J Nutr. 2000;83(Suppl 1):S67–70. 45. Marshall JA, Hamman RF, Baxter J, et al. Ethnic differences in risk factors associated with the prevalence of noninsulin-dependent diabetes mellitus. The San Luis Valley Diabetes Study. Am J Epidemiol. 1993;137:706 –18. 46. Bjorntorp P, Rosmond R. Visceral obesity and diabetes. Drugs. 1999;58(suppl):13– 8. 47. Rosmond R, Holm G, Bjorntorp P. Hypothalamic arousal, insulin resistance, and type 2 diabetes mellitus. Diabet Med. 1999;16:373– 83. 48. Salans LB, Knittle JL, Hirsch J. The role of adipose cell size and adipose tissue insulin sensitivity in the carbohydrate intolerance of human obesity. J Clin Invest. 1968;47:153– 65. 49. Bjorntorp P. Fatty acids, hyperinsulinemia, and insulin resistance: which comes first? Curr Opin Lipidol. 1994;5:166 –74. 50. Evans DJ, Murray R, Kissebah AH. Relationship between skeletal muscle insulin resistance, insulin-mediated glucose disposal, and insulin binding: effects of obesity and body fat topography. J Clin Invest. 1984;74:1515–25. 51. Dowling HJ, Fried SK, Pi-Sunyer FX. Insulin resistance in adipocytes of obese women: effects of body fat distribution and race. Metabolism. 1995;44:987–95. 52. Landis JR, Lepkowski JM, Eklund SA, Stehouwer SA. A statistical methodology for analyzing data from a complex survey: the first National Health and Nutrition Examination Survey. Vital Health Stat 2. 1982;92:1–52. 53. Fagard RH. Physical activity in the prevention and treatment of hypertension in the obese. Med Sci Sports Exerc. 1999; 31(Suppl 11):S624 –30. 54. Esparza J, Fox C, Harper IT, et al. Daily energy expenditure in Mexican and USA Pima indians: low physical activity as a possible cause of obesity. Int J Obes Relat Metab Disord. 2000;24:55–9.


9