Hyperglycemia, Hyperinsulinemia, Overweight, and High Blood Pressure in Young Adults
The Rio de Janeiro Study
Abstract Children and adolescents (n=3906, 10-15 years old) have been participating in a screening program for high blood pressure. Sixty-four individuals (17-23 years old) from this population were followed up for 8 years and four consecutive screenings and were stratified into three groups according to blood pressure: group 1 (n=23), ≥95th percentile for at least three of four evaluations; group 2 (n=28), <50th percentile for at least three of four screenings; and group 3 (n=13), with unstable blood pressure percentiles. All 64 individuals underwent an oral glucose tolerance test after a 12-hour fast. Blood samples were collected at 0, 30, 60, 90, and 120 minutes for insulin and glucose measurements. Group 1 had a greater body mass index and higher systolic and diastolic blood pressures, basal glucose and insulin levels, and peak values of insulin and glucose levels than the other groups (P<.05). Group 1 also had a higher prevalence of overweight and abnormal values of basal insulin than the other groups (P<.05) and a higher proportion of glucose-intolerant individuals when compared with group 2 (P<.05). Systolic and diastolic blood pressures were positively related to body mass index (P<.05) and insulin variables (P<.05); however, when body mass index was controlled for, only systolic blood pressure demonstrated a significant correlation with insulin variables (P<.05). The association of overweight, hyperinsulinemia, glucose intolerance, and high blood pressure can be detected early, but the significance of these findings would be better explained by longitudinal studies.
Abnormalities in carbohydrate metabolism have been associated with increased mortality from cardiovascular disease.1 2 3 Several studies have described an epidemiological association among essential hypertension, glucose intolerance, dyslipidemia, and central obesity, suggesting a possible common pathogenetic mechanism related to hyperinsulinemia and peripheral insulin resistance.4 5 6 Hyperinsulinemia and insulin resistance are present in almost all obese or type II diabetic individuals with hypertension6 and also in 50% of nonobese hypertensive patients.7 Moreover, prospective studies have demonstrated that in middle-aged whites and Mexican-American individuals, fasting insulin levels are closely related to the future development of hypertension, even when adjustments are made for age and weight.1
The relationship among blood pressure (BP), central obesity, and high levels of insulin observed in adults has also been detected in children and adolescents.8 9 The Bogalusa Heart Study has reported that fasting glucose and insulin levels are related to blood pressure, body size, and cardiovascular risk factors in individuals aged 5 to 17 years old.8 In view of the above considerations, the aim of this study was to evaluate early changes in glucose and insulin response after ingestion of a glucose load and to correlate these changes with overweight and blood pressure in young adults with different percentiles of blood pressure.
Sixty-four individuals were selected for the current analysis from a larger cohort study: 3906 children and adolescents from 10 to 15 years of age have been participating in a screening program for arterial BP determination in the city of Rio de Janeiro since 1987. Normal BP curves were established according to age and sex, and this population was separated into two groups according to BP percentile: group 1 included 327 subjects with systolic (SBP) and/or diastolic BP (DBP) at or above the 95th percentile, and group 2 included 327 subjects randomly selected from those whose SBP and DBP were at or below the 50th percentile as a control group (phase 1). From these two groups, 206 individuals (106 from group 1 and 100 from group 2) were visited at home from 1989 to 1990, at which time their parents and siblings were also examined (phase 2) for blood pressure. For medical evaluation of the possible effects of BP on target organs, 127 children and adolescents (70 from group 1 and 57 from group 2) who participated in phase 2 agreed to be examined at the hospital during 1991 to 1992 (phase 3). From this population, 64 subjects (28 from group 1 and 36 from group 2), age 17 to 23 years, underwent an oral glucose tolerance test (OGTT) during 1994 to 1995 to evaluate glucose and insulin responses (phase 4). These four consecutive screening periods were able to identify 13 individuals (7 from group 1 and 6 from group 2) who did not maintain the same BP percentiles for at least three of four BP measurements. According to this finding, the sample was restructured into three groups: group 1 included 23 subjects (12 males and 11 females) who maintained an SBP and/or a DBP ≥95th percentile for at least three of four evaluations; group 2 included 28 subjects (13 males and 15 females) who maintained an SBP and DBP <50th percentile for at least three of four evaluations; and group 3 included 13 subjects (3 males and 10 females) who did not maintain the same BP percentile for at least three of four evaluations. All subjects agreed to participate in the study and signed a consent form.
The anthropometric indexes used were weight (kilograms), height (meters), and body mass index (BMI=weight [kg]/height2 [m2]). Overweight was defined as a BMI ≥25 kg/m2.
In all phases, BP was measured in the supine position using a wall-mounted or table mercury type sphygmomanometer (Tycos) on the right arm, with cuff sizes of 7.5, 9.5, 12, and 14 cm in width and 36 and 53 cm in length according to American Heart Association recommendations.10 Three BP measurements were obtained and the value of the last determination was used for analysis. Diastolic BP was determined at Korotkoff phase V.
The 95th and 50th BP percentile values used for stratification are shown in Table 1⇓. These values were obtained from phase 1 of the Rio de Janeiro Study. The age-15 values were used for classifying BP at ages 16 and 17. For ages equal or above 18, 130/85 mm Hg was used as the cut point to determine abnormal values for systolic/diastolic BP.
The OGTT was performed using 75 g of Dextrosol after a 12-hour fast. Venous blood samples were collected in siliconized tubes at 0, 30, 60, 90, and 120 minutes, and after blood coagulation, the samples were centrifuged at 3000 rpm for 15 minutes. Glucose levels were immediately determined and insulin levels were analyzed later from serum samples stored at −20°C for a maximum period of 3 months.
Insulin levels were measured by radioimmunoassay (Diagnostic Products Corp), and plasma glucose levels were determined in a Cobas Miras Plus automatic Roche analyzer by a colorimetric enzymatic method.
The variables analyzed in the present study were the following: baseline levels of insulin (BI) and glucose (BG) and peak values of insulin (PVI) and glucose (PVG), defined as the highest value after the glucose overload. According to the Diagnostic Products Corp assay, a BI ≥30 μIU/mL was considered abnormal. Glucose intolerance (GI) was defined as a glucose level between 140 and 200 mg/dL 2 hours after the ingestion of 75 g of Dextrosol (World Health Organization criteria11 ).
Descriptive data are presented as mean±SD (range). Statistical analysis was performed with ANOVA (F), complemented by Tukey’s honestly significant difference test, Kruskal-Wallis one-way ANOVA by ranks (H), complemented by a nonparametric multiple-comparison test, χ2 test, multiple regression, and correlation. Insulin variables were logarithmically transformed prior to correlation analysis.
Age and sex were not different among the groups, which allowed us to study each group as a whole. Weight and BMI were significantly higher in group 1; however, no difference was found in height. SBP, DBP, BI, BG, PVI, and PVG were statistically higher in group 1 when compared with the other groups (Table 2⇓).
Fig 1⇓ shows the insulin and glucose curves obtained for the three study groups. Group 1 had higher levels of glucose at baseline and at 30 and 120 minutes after Dextrosol ingestion than the other groups. Group 1 also had higher insulin levels at 0 (baseline), 30, and 90 minutes after glucose overload.
Group 1 had a higher prevalence of overweight (7 [30.4%] from group 1, 3 [10.7%] from group 2, and 0 from group 3 [χ2=6.75, P=.03]) and a higher prevalence of abnormal BI values (5 [21.7%] from group 1, 1 [3.6%] from group 2, and 0 from group 3 [χ2=6.59, P=.04]) when compared with the other two groups. The presence of GI did not differ among the groups: 3 (13.1%) in group 1, 0 in group 2, and 0 in group 3 (χ2=5.61, P=.06); however, a statistically significant difference was found when group 1 was compared with group 2 (χ2=3,88, P=.04).
SBP was positively related to BMI and insulin and glucose variables (BI, PVI, BG, and PVG); however, DBP demonstrated significant correlation only with BMI and glucose variables (BG and PVG) as shown in Table 3⇓. When BMI was controlled for, only BI, PVI, and PVG maintained significant correlations with SBP. The partial correlation coefficients were .27 (P=.0384) for BI, .33 (P=.0069) for PVI, .22 (P=.0891) for BG, and .36 (P=.0024) for PVG. In a similar model, only PVI maintained a significant correlation with DBP when BMI was controlled for; these partial correlation coefficients were .17 (P=1977) for BI, .47 (P=.0001) for PVI, .14 (P=.2747) for BG, and .2008 (P=.1000) for PVG.
Arterial hypertension occurs frequently in association with visceral obesity and diabetes mellitus or glucose intolerance, and it is now accepted that all of these clinical conditions are part of a metabolic syndrome that also includes specific alterations in lipid metabolism and atherosclerotic cardiovascular disease.4 5 6 The observation that a great majority of hypertensive patients are hyperinsulinemic has led investigators to look for a possible involvement of insulin in the development of hypertension. Nevertheless, the mechanism of hyperinsulinemia and insulin resistance in the pathogenesis of essential hypertension is not completely understood. Different pathways have been proposed, including an increase in kidney sodium and water reabsorption,12 13 the enhancement of sympathetic nervous system activity,14 alterations in transmembrane electrolyte transport leading to intracellular Ca2+ accumulation,15 and direct or indirect stimulation of cellular proliferation in the arterial wall.16
On the other hand, experimental studies in dogs failed to demonstrate a relationship between hyperinsulinemia and hypertension.17 In human beings, it was observed that individuals with insulinoma did not develop high blood pressure levels.18 Despite the fact that hyperinsulinemia and insulin resistance are found more frequently among hypertensive obese and/or type II diabetic individuals, ≈50% of nonobese hypertensive subjects can be hyperinsulinemic.7
Haffner et al,1 in a prospective study of 8 years, found that individuals who were nondiabetic at the time of the baseline examination but who subsequently developed type II diabetes had higher levels of total and LDL cholesterol, triglyceride, fasting glucose and insulin, BMI, and blood pressure and lower levels of HDL cholesterol than did those subjects who remained nondiabetic. In young populations, there are no similar available studies, although some reports have shown a close correlation of serum lipids and insulin in children and adolescents.8 19 The Bogalusa Heart Study found a strong correlation between fasting insulin and serum lipids and lipoproteins, especially triglyceride and VLDL cholesterol levels, that were independent of age and weight.8
In our study, hyperinsulinemia was present in the group of young individuals with persistent upper percentiles of BP, as shown by high baseline and peak value levels of insulin after the glucose overload. This finding suggests insulin resistance, although only a euglycemic-hyperinsulinemic clamp can confirm it.
SBP and DBP were closely related to BMI and fasting and peak levels of insulin. Fasting and peak levels of insulin maintained a significant correlation with SBP, even when BMI was controlled for. The association of insulin levels, obesity, and high BP has been shown in children and adolescents. The Bogalusa Heart Study, which evaluated 3313 children and adolescents aged 5 to 17 years, found that fasting insulin levels were positively related to some anthropometric indexes and to systolic and diastolic blood pressure measurements.8 Voors et al20 observed that different measures of obesity were highly related to insulin response.
Our data also show that young adults with persistent high BP percentiles have a higher prevalence of overweight and glucose intolerance. In addition, basal and peak levels of glucose were positively related to SBP. Other investigators have observed similar relationships between glucose and BP. The Bogalusa Heart Study showed that fasting glucose levels were positively correlated with systolic and diastolic BP.8 Florey et al21 also demonstrated an association between glucose and BP levels in 2388 children aged 9 to 12 years.
Glucose intolerance is a common finding in hypertensive subjects.22 Modan et al,5 in an epidemiological study performed in Israel, found that 52.9% of hypertensive individuals presented with impaired glucose tolerance. Harris23 demonstrated that the risk of development of hypertension is twice as great in glucose-intolerant than glucose-tolerant individuals. These findings, however, were obtained from middle-aged populations, and no prospective data are available for young subjects.
The present study provides evidence that hyperinsulinemia may be detected early in young adults with persistent elevated BP percentiles and a high prevalence of overweight. These findings may suggest that hyperinsulinemia in young adults may play a role in the development of arterial hypertension or may be part of a syndrome that could be diagnosed in the future.
In conclusion, we found that young individuals with persistent upper percentiles of blood pressure may also have a cluster of risk factors for the future development of coronary artery disease, such as impaired glucose tolerance, hyperinsulinemia, and overweight. Moreover, these findings emphasize the importance of a careful follow-up of this group and implementation of a preventive approach to disease.
Selected Abbreviations and Acronyms
|BG||=||basal glucose level|
|BI||=||basal insulin level|
|BMI||=||body mass index|
|DBP||=||diastolic blood pressure|
|log BI||=||basal insulin log transformed|
|log PVI||=||peak value of insulin log transformed|
|PVG||=||peak value of glucose|
|PVI||=||peak value of insulin|
|SBP||=||systolic blood pressure|
- Received March 17, 1997.
- Revision received April 17, 1997.
- Accepted May 7, 1997.
Kannel WB, McGee DI. Diabetes and glucose tolerance as risk factors for cardiovascular disease: the Framingham Study. Diabetes Care. 1979;2:120-126.
Stamler J, Vaccaro O, Neaton JD, Wentworth D, for the Multiple Risk Factor Intervention Trial Research Group. Diabetes, other risk factors and 12 year cardiovascular mortality for men screened in the Multiple Risk Factor Intervention Trial. Diabetes Care. 1993;16:434-444.
Simonson DC. Etiology and prevalence of hypertension in diabetic patients. Diabetes Care. 1988;11:821-827.
de Fronzo RA, Ferranini E. Insulin resistance: a multifaceted syndrome responsible for NIDDM, obesity, hypertension, dyslipidemia and atherosclerotic cardiovascular disease. Diabetes Care. 1991;14:173-194.
Kaplan, NM. Primary ypertension: pathogenesis. In: Kaplan NM, ed. Clinical Hypertension. 6th ed. Baltimore, Md: Williams & Wilkins; 1994:47-108.
World Health Organization. Diabetes Mellitus. Report of WHO Study Group. Geneva, Switzerland: World Health Organization; technical report series. 1985:727.
Gupta AK, Clark RV, Kirchner KA. Effects of insulin on renal sodium excretion. Hypertension. 1992;19(suppl I):1-78-1-82.
Daly PA, Landsberg L. Hypertension in obesity and NIDDM: role of insulin and sympathetic nervous system. Diabetes Care. 1991;14:240-248.
Mulvaney MJ. Pathophysiology of vascular smooth muscle in hypertension. J Hypertens. 1984;2(suppl III):413-420.
Hall JE, Brands MW, Dixon WM. Hyperinsulinemia does not elevated blood pressure in obese hypertensive dogs. FASEB J. 1991;5:A737. Abstract.
Islam AH, Yamashita S, Kotani K, Nakamura T, Tokunaga K, Arai T, Nishida M, Kameda-Takemura K, Matsuzawa Y. Fasting plasma insulin level as an important risk factor for the development of complications in Japanese obese children: results from a cross-sectional and longitudinal study. Metabolism. 1995;44:478-485.
Voors AW, Harsha DW, Webber LS. Clustering of anthropometric parameters, glucose tolerance, and serum lipids in children with high and low β- and pre-β-lipoproteins: Bogalusa Heart Study. Atherosclerosis. 1982;2:346-355.
Forey C du V, Uppal S, Lowy C. Relation between blood pressure, weight, and plasma sugar and serum insulin levels in school children aged 9-12 years in Westland, Holland. Br Med J. 1976:1:1368-1371.
Harris MI. Impaired glucose tolerance in the US population. Diabetes Care. 1989;12:464-474.