AJH
2005; 18:3–12
Original Contributions
Metabolic Syndrome and Insulin Resistance in
the TROPHY Sub-Study: Contrasting Views in
Patients With High-Normal Blood Pressure
Brent M. Egan, Vasilios Papademetriou, Marion Wofford, David Calhoun,
Jyotika Fernandes, Jessica E. Riehle, Shawna Nesbitt, Eric Michelson,
Stevo Julius, for the TROPHY Sub-study Investigative Team
Background: Although insulin resistance and metabolic syndrome are often used synonymously, concordance is not established.
Methods: Metabolic, hemodynamic, and hormonal
data were analyzed on 141 patients in the Trial of Preventing Hypertension (TROPHY) Sub-Study with highnormal blood pressure (BP) (130 to 139/85 to 89 mm Hg
[mean ⫾ SD, 133 ⫾ 8/85 ⫾ 6 mm Hg]; age, 48 ⫾ 9 years;
body mass index 30 ⫾ 5 kg/m2).
Results: Fifty-three of 141 subjects (37.6%; ⬃3/8) had
the metabolic syndrome based on three or more of the five
risk factors (BP, waist circumference, fasting triglycerides,
HDL-cholesterol, glucose). To maintain consistency in proportions, insulin resistance was defined as the upper 3/8 of the
distribution on the homeostatic model assessment (HOMA),
which uses fasting glucose and insulin and a modified Matsuda-DeFronzo index, based on fasting, 1- and 2-h glucose
and insulin values. Among metabolic syndrome patients,
57% and 55% were in the upper 3/8 of the distribution for
insulin resistance by HOMA and Matsuda-DeFronzo, respectively. Among subjects without the metabolic syndrome,
26% and 27% were insulin resistant by HOMA and MatsudaDeFronzo criteria. The proportion of patients with metabolic
syndrome and insulin resistance increased strongly and similarly with increasing body mass index. However, metabolic
syndrome and insulin resistance were different compared
with their respective controls in the lower 5/8 of the distribution, in waist/hip ratios, fasting and 1-h insulin, HDLcholesterol, heart rate, and systolic BP responses to exercise
and plasma renin, angiotensin, and aldosterone.
Conclusions: The findings suggest that metabolic syndrome and insulin resistance are not synonymous anthropometrically, metabolically, hemodynamically, or hormonally
in patients with high-normal BP. Am J Hypertens 2005;18:
3–12 © 2005 American Journal of Hypertension, Ltd.
Key Words: Metabolic syndrome, insulin resistance,
blood pressure, arterial compliance, exercise, leptin, renin,
angiotensin.
lustering of cardiovascular risk factors and excess
cardiovascular and renal events associated with
obesity have been recognized for more than 80
years.1 The worldwide obesity epidemic is driving an
increasing burden of preventable cardiovascular risk and
disease.2,3 Clinical investigators have defined the constellation of risk factors largely from the perspective of hyperinsulinemia and insulin resistance.4 – 6 However, the
complex methodology required to quantify insulin resis-
C
tance and the lack of a standardized insulin assay have
impeded the translation of research findings on insulin
resistance into clinical practice. The Adult Treatment
Panel (ATP-III) addressed this void with diagnostic criteria for the metabolic syndrome based on three or more of
the five clinical variables, including BP, waist circumference, and fasting values for HDL-cholesterol, triglycerides, and glucose.7 The metabolic syndrome and insulin
resistance are both strongly related to body mass index,6,8
Received March 31, 2004. First decision August 4, 2004. Accepted
August 4, 2004.
From the Medical University of South Carolina (BME, JF, JER), Charleston,
South Carolina; Georgetown University (VP), Washington, DC; University of Mississippi (MW), Jackson, Mississippi; University of Alabama
(DC), Birmingham, Alabama; University of Texas Southwestern (SN),
Dallas, Texas; AstraZeneca Pharmaceuticals LP (EM), Wilmington, Delaware; and University of Michigan (SJ), Ann Arbor, Michigan.
The data in this report reflect baseline information obtained in a
subset of patients participating in the Trial of Preventing Hypertension
(TROPHY) sponsored by AstraZeneca Pharmaceuticals LP, Wilmington
DE. The TROPHY Sub-study was also supported in part by the National
Institutes of Health (NIH) Heart Lung & Blood Institute HL58794,
HL04290 and by General Clinical Research Center RR-01070 from the
NIH, Division of Research Resources.
Address correspondence and reprint requests to Dr. Brent M. Egan,
Medical University of South Carolina, 96 Jonathan Lucas Street, CSB
826H, Charleston, SC 29425; e-mail: [email protected]
© 2005 by the American Journal of Hypertension, Ltd.
Published by Elsevier Inc.
0895-7061/05/$30.00
doi:10.1016/j.amjhyper.2004.08.008
4
METABOLIC SYNDROME AND INSULIN RESISTANCE
AJH–January 2005–VOL. 18, NO. 1
which suggests the terms may be interchangeable. The
magnitude of overlap between the metabolic syndrome
and various definitions of insulin resistance are not well
documented and could be important in the effort to translate clinical– epidemiological research findings into medical practice.
Blood pressure (BP) in the high-normal range identifies
individuals more likely to develop hypertension9 and to
have cardiovascular risk factor clustering.10,11 The Trial of
Preventing Hypertension (TROPHY) enrolled 809 patients
with high-normal BP to determine whether randomized
assignment to the angiotensin receptor blocker candesartan versus placebo for the first 2 years reduces new onset
hypertension during the 4-year study.12,13 The TROPHY
Sub-study collected additional baseline data on metabolic,
hemodynamic, and neurohormonal variables to better understand pathophysiology, to further elucidate the physiological response to angiotensin receptor blockade, and to
refine predictors of progression to hypertension.
This TROPHY Sub-study report focuses on the congruity between the metabolic syndrome defined by ATPIII and two clinically applicable indices for quantifying
insulin resistance.7,14,15 A secondary objective is to assess
the concordance between insulin resistance and metabolic
syndrome on anthropometric, metabolic, neurohormonal,
and hemodynamic data.
of supine rest using HDI / Pulsewave (Hypertension Diagnostics Inc., Eagan, MN).17
Heart rate in the office was determined by the Omron
device, by the HDI / Pulsewave in the laboratory, and by
palpation of the radial pulse immediately before, during,
and after the standardized treadmill test.
Blood pressure and heart rate were measured in triplicate using mercury sphygmomanomtry and appropriately
sized arm cuff after 5 min of rest in the seated position
with single measurements after 2 min standing quietly and
then at 3 (stage 1, 1.7 miles/h at 10% elevation) and 6 min
(stage 2, 2.5 miles/h at 12% elevation) on a treadmill
(modified Bruce protocol).18 Postexercise BP and heart
rate were obtained with volunteers sitting.
Methods
Study Design
The objectives and methods for TROPHY were described.12,13
Volunteers
All subjects in this report provided informed consent approved by the Institutional Review Boards of each institution.
Eligible patients included men and women 30 to 65 years old
with systolic or diastolic BP on three consecutive visits in the
high-normal range (130 to 139/85 to 89 mm Hg).16 Subjects
with treated hypertension in the previous 6 months or any
history of cardiovascular disease were excluded. Patients
requiring treatment for diabetes mellitus, long-term treatment
for any condition with medications that affect BP, serum
creatinine ⬎1.5 mg/dL or serum potassium ⬍3.4 mEq/L
were excluded. Estrogen replacement therapy and intermittent use of nonsteroidal anti-inflammatory agents and
decongestants were permitted except within 24-h of BP measurements.
Measurements
Hemodynamic The principal BP measurement was
based on the mean of three readings obtained with Omron
HEM-705CP (Vernon Hills, IL) automated monitor beginning after 5 min of rest with subjects seated. Blood pressure during the laboratory study was obtained after 15 min
Anthropometric Variables With subjects standing,
waist (abdominal) circumference was measured over the skin
to the nearest 0.1 cm at the maximum anterior extension of
the abdomen using an inelastic tape.
With subjects standing, hip circumference was measured over an undergarment to the nearest 0.1 cm at the
maximum posterior extension of the buttocks.
Metabolic Data A 75-g (Sun-Dex 75, Fisherbrand, Suwanee, GA), 2-h oral glucose tolerance test was used to
assess insulin resistance.14,15,19 Insulin resistance was defined
by modification of the Matsuda-DeFronzo index (10,000/
square root [fasting glucose ⫻ fasting insulin] ⫻ [mean
glucose ⫻ mean insulin during the oral glucose tolerance
test]).14 Insulin resistance was also determined using homeostasis model assessment (HOMA) ⫽ (fasting insulin [in
microunits per milliliter] ⫻ fasting glucose [in millimoles per
liter]/22.5).15 Both indices correlate strongly with the euglycemic clamp.15
Fasting lipids and lipoproteins were assayed using established methods by Covance, Inc., Princeton, NJ.
Neurohormonal Variables Plasma norepinephrine was
measured by high performance liquid chrormotography
with electrochemical detection after extraction with aluminum oxide. The coefficient of variation for norepinephrine ranged from 8.6% to 9.6%.
The gamma goat 125I plasma renin activity radioimmunoassay kit (DiaSorin, Stillwater, MN) kit was used. The
sensitivity of the assay is 0.2 ng/mL/h angiotensin I. The
intra-assay coefficients of variation were 5.0% to 5.9% and
the interassay ranges 4.6% to 10.0%.
The double-antibody immunoreactive assay methodology was used to measure plasma angiotensin II.20 The
sensitivity of the assay was 0.7 pg/mL.
The Diagnostics Products Corporation (Los Angeles,
CA) Coat-A-Count radioimmunoassay kit was used to
measure plasma aldosterone. The sensitivity of the assay is
1.6 ng/dL. The intra-assay and interassay coefficients of
variation ranged from 2.7% to 8.7% and 3.9% to 10.4%,
respectively.
The Abbott (Abbott Park, IL) IMX Microparticle Enzyme Immunoassay was used to measure serum insulin.
5
AJH–January 2005–VOL. 18, NO. 1
METABOLIC SYNDROME AND INSULIN RESISTANCE
Intra-assay and interassay coefficients of variation ranged
from 2.3% to 3.0% and 3.7% to 4.9%, respectively.
The DRG Leptin ELISA (DRG Instruments, GMbH, Germany) assay was used to measure plasma leptin. Sensitivity
of the assay is 0.49 ng/mL with intra-assay and inter-assay
coefficients of variation of 11.0% to 15.5% and 12.3% to
16.2%, respectively. Samples with values ⬎25 ng/mL were
diluted and repeated and the result multiplied by the dilution
factor.
or insulin resistance by the two indices and their respective
controls (lower 5/8 or 62.5% of the distribution) were
assessed using two-sample Student t test, ␹2 test, or repeated measures analysis of variance (ANOVA) as appropriate. Two-tailed tests were used except where indicated
and when previous studies demonstrated directionally consistent differences. P values ⬍ .05 were accepted as significant.
Protocol
Patients in the TROPHY Sub-study completed the main
study measurements.12,13 Sub-study volunteers reported to
the clinical laboratory in the morning after an overnight
fast. Blood pressure was measured in triplicate after 5 min
of rest in the sitting position using the Omron. Patients
assumed the recumbent position, and a heparin-lock was
placed. Thirty minutes later, blood was drawn for catecholamines, renin, angiotensin, aldosterone, leptin, glucose, and insulin. Subjects drank the Sun-dex 75. Blood
was drawn 60 and 120 min later for glucose and insulin,
and the heparin-lock was withdrawn. After a light meal,
patients underwent measurements of waist and hip circumferences and biceps, triceps, subscapular, and suprailiac
skinfold thicknesses. The exercise test was performed.
Results
Baseline Characteristics of the TROPHY
Sub-study Cohort
Baseline demographic, anthropometric, hemodynamic, metabolic, and neurohormonal characteristics of the TROPHY
Sub-study cohort are provided in Table 1. Sub-study volunteers were on average overweight and obese middle-aged
adults with a high proportion of white men.
Metabolic Syndrome Risk Factors
All study volunteers had at least one criterion for the
metabolic syndrome, that is, a BP ⱖ130 mm Hg systolic
or ⱖ85 mm Hg diastolic. As shown in Fig. 1, 88 of 141
subjects (62.4%, ⬃5/8) had one or two metabolic syndrome risk factors,7 whereas 53 of 141 (37.6%, ⬃3/8) had
three or four risk factors.
Data Management and Analysis
Analytical approaches in the main TROPHY study were
reported.12,13 Case report forms and flow sheets for the
Sub-study protocol were completed by each site and forwarded to the data coordinating center at the University of
Michigan (Ann Arbor, MI). Laboratory samples were sent
to Covance Inc., (Princeton, NJ), and results were forwarded to the data coordinating center. Selected data from
the main TROPHY study required for this Sub-study report were transferred from AstraZeneca (Wilmington, DE)
to Ann Arbor. Data for each Sub-study subject were
merged in the Microsoft Excel database and transferred to
the Medical University of South Carolina for analysis.
Patients were stratified by metabolic syndrome7 and insulin resistance criteria.14,15 Metabolic syndrome was defined
by three or more of the five risk factors (BP ⱖ130 mm Hg
systolic or ⱖ85 mm Hg diastolic, waist circumference ⬎102
cm for men or ⬎88 cm for women, fasting glucose ⱖ110
mg/dL, triglycerides ⬎150 mg/dL, or HDL-cholesterol ⬍50
mg/dL for women or ⬍40 mg/dL for men).7 Descriptive
analyses were generated for anthropometric, metabolic, hemodynamic, and neurohormoanal variables.
Because 53 of 141 subjects (37.6% or 3/8) met the
metabolic syndrome definition, the same proportion (upper
3/8 or ⬃37.5%) was used to define insulin resistance by
HOMA and Matsuda-DeFronzo criteria. Insulin resistance
was not defined using cut-points from previous studies, as
the insulin assay has not been standardized across laboratories and results are not directly comparable.
Differences between groups with metabolic syndrome
Insulin Resistance and Concordance with
Metabolic Syndrome
Metabolic syndrome subjects were more likely to occupy
the same relative rank by HOMA or Matsuda-DeFronzo
(Fig. 1). A significant minority of metabolic syndrome
patients was not among the insulin resistant group and vice
versa.14,15
Concordance between indices of insulin resistance was
greater than concordance between insulin resistance and metabolic syndrome. The HOMA and Matsuda-DeFronzo indices were concordant in 89% (N ⫽ 116) of patients (58%,
lower 5/8 of the distribution for both; 31%, upper 3/8 of the
distribution for both) and discordant in 11% (N ⫽ 13).
Subjects with two and three metabolic syndrome risk factors
were not different in the proportions with insulin resistance (P
⬎ .60).
The distribution of body mass indices in the lean,
overweight, and obese categories among patients with the
metabolic syndrome and insulin resistance is shown on the
left side of Fig. 2; the percentage of lean, overweight, and
obese subjects with high-normal BP who had the metabolic syndrome and insulin resistance are depicted on the
right side.
Table 2 provides comparative data for volunteers with
the metabolic syndrome or insulin resistance (HOMA and
Matsuda-DeFronzo), compared to respective controls.
Comparisons between metabolic syndrome or insulinresistant subjects and their respective controls are shown
in Table 3. All three groups, compared to their controls,
6
METABOLIC SYNDROME AND INSULIN RESISTANCE
Table 1. Characteristics of patients with high-normal BP in the TROPHY sub-study
Characteristic
Demography and health habits
Age
Sex (male: female)
Ethnicity (C:AA:H)
Current smokers
Alcohol (ⱖ1 drink/wk)
Anthropometric variables
BMI (kg/m2)
Abdominal circumference
(inches)
Hip circumference (inches)
Abdominal:hip circumference
ratio
Baseline hemodynamic variables
Systolic BP (mm Hg)
Diastolic BP (mm Hg)
Heart rate (beats/min)
Baseline metabolic variables
Cholesterol (mg/dL)
Triglycerides (mg/dL)
HDL-cholesterol (mg/dL)
LDL-cholesterol (mg/dL)
Fasting glucose (mg/dL)
Fasting insulin, (␮U/mL)
Baseline neurohormonal
variables
Plasma leptin, (ng/mL)
Plasma norepinephrine (pg/mL)
Plasma renin activity (AI/mL/h)
Plasma angiotensin II (pg/mL)
Plasma aldosterone (ng/dL)
Mean ⴞ SD or
N, %
48.4 ⫾ 8.5
98:45 (69.5%:
30.5%)
112:23:2,
(81.8%:16.8%)
21 (14%)
85 (62%)
29.6 ⫾ 5.8
38.7 ⫾ 6.1
44.9 ⫾ 15.0
0.90 ⫾ 0.17
133 ⫾ 8
85 ⫾ 6
75 ⫾ 11
193
136
47
119
92
9.6
⫾
⫾
⫾
⫾
⫾
⫾
35
87
14
34
10
7.9
14.5
264
0.89
5.5
7.2
⫾
⫾
⫾
⫾
⫾
17.2
128
0.94
3.2
4.6
AA ⫽ African American; BMI ⫽ body mass index; BP ⫽ blood pressure; C ⫽ white; H ⫽ Hispanic.
were distinguished anthropometrically by greater weight,
body mass index, and waist circumference, metabolically
by elevated fasting, 1- and 2-h glucose, fasting insulin and
triglycerides, and a strong tendency to higher total cholesterol/HDL ratios, and hemodynamically by greater systolic
BP at 6 min of exercise. Differences between metabolic
syndrome and insulin-resistant groups and their respective
controls were less consistent for waist/hip ratios, heart
rates at rest and during exercise, HDL-cholesterol, fasting
and 1-h insulin, and markers of renin-angiotensin-aldosterone system activity.
Glucose and insulin at baseline and during the glucose
tolerance test for subjects stratified by metabolic syndrome
and insulin resistance are displayed in Fig. 3. As noted,
patients with the syndrome did not have higher insulin at
baseline or at 1 h than patients without.
Fig. 4 shows systolic BP and heart rate at baseline and
during treadmill exercise. Separation was more apparent
visually and statistically for subjects stratified by insulin
resistance than by metabolic syndrome.
AJH–January 2005–VOL. 18, NO. 1
Discussion
Translation of research findings on insulin resistance and
the metabolic syndrome into medical practice has been
hampered by the lack of a practical clinical definition. The
ATP-III definition of the metabolic syndrome represents a
constructive response to the gap, as it uses five readily
obtained clinical variables including BP, waist circumference, fasting glucose, triglycerides, and HDL-cholesterol.7
Despite the implied agreement between the metabolic syndrome and insulin resistance, the degree of concordance
has not been established.
Metabolic Syndrome Prevalence
Among patients with high-normal BP in the TROPHY
Sub-Study, 53 of 141 (37.6% or ⬃3/8) met metabolic
syndrome criteria (Fig. 1), which is 50% higher than the
prevalence of ⬃1/4 in the US population.7 The higher
prevalence could simply reflect that all individuals began
with one risk factor, that is, high-normal BP. The median
age for this predominantly white group is slightly older
than that for whites in NHANES III (45 years for men,
47.6 for women), which would also contribute to a higher
rate of metabolic syndrome.8 However, patients with highnormal BP have more metabolic risk factors.11–13 And,
groups with a high probability of the metabolic syndrome,
for example, those with hypertension, cardiovascular disease, diabetes mellitus requiring pharmacologic treatment,
and renal insufficiency were excluded from our study. The
fact that patients with high-normal BP are more likely than
the general population to have the metabolic syndrome,
which is predictive of coronary outcomes and new onset
diabetes mellitus,21,22 is clinically relevant.
Similarities and Differences Between the
Metabolic Syndrome and Insulin
Resistance
Anthropometric Characteristics The proportion of patients defined as insulin resistant increased as a function of
metabolic syndrome risk factors (Fig. 1) and body mass index
(Fig. 2). The metabolic syndrome and insulin resistance are
both strongly related to BMI (Fig. 2).6,8 In NHANES III,
4.6%, 22.4%, and 59.6% of men and 6.2%, 28.1%, and
50.0% of women who were normal, overweight, and obese,
respectively, had the metabolic syndrome.8 The relationship
of body mass index to the metabolic syndrome is similar for
lean and obese subjects in our study and the previous report.8
Overweight subjects with high-normal BP in our study appear more likely to have the metabolic syndrome than overweight adults in general.7,8
Metabolic syndrome patients had higher waist circumferences than those without, but they did not have significantly
greater waist (abdominal)/hip ratios (Tables 2 and 3). Waist/
hip ratios were significantly different when stratifying subjects by insulin resistance. Patients with higher waist/hip
ratios are more likely to have hyperinsulinemia, insulin re-
AJH–January 2005–VOL. 18, NO. 1
FIG. 1 (Top) Distribution of metabolic syndrome risk factors
among 141 patients with high-normal blood pressure in the TROPHY
substudy. (Bottom) Percentages of patients in the upper 3/8 of
insulin resistance by either the HOMA (N ⫽ 128) or Matsuda index (N
⫽ 119) stratified by the number of metabolic syndrome risk factors.
sistance, dyslipidemia, and high BP.4,23 Our findings are
consistent with other evidence that abdominal obesity defined
by waist circumference alone is a cardiovascular disease risk
factor.23,24
METABOLIC SYNDROME AND INSULIN RESISTANCE
7
Metabolic Variables Patients with the metabolic syndrome are presumed to be hyperinsulinemic, and fasting
insulin is included in the European definition of metabolic
syndrome.21,25 In this study, patients with the metabolic
syndrome did not have higher fasting or 1-h post-glucose
insulin values. Patients with the metabolic syndrome are also
regarded as insulin resistant. Patients with four metabolic
syndrome risk factors were the most insulin-resistant group,
whereas individuals with only a high-normal BP were the
least insulin resistant. Yet the proportion of patients with two
and three metabolic syndrome risk factors who were also
insulin resistant was comparable (Fig. 1). Patients with highnormal BP and only one other metabolic syndrome risk factor
may be at significant risk. Among 1108 patients undergoing
cardiac catheterization, those with one to four metabolic
syndrome risk factors had comparable proportions with
⬎50% stenosis, history of myocardial infarction, and coronary intervention including coronary bypass.26 In the absence
of compelling treatment indications and a metabolic syndrome designation, some patients with high-normal BP may
be falsely considered low risk.
Subjects with the metabolic syndrome had higher triglycerides and lower HDL-cholesterol and lower total/HDLcholesterol ratios than subjects without the syndrome, as
expected.7 However, these lipid abnormalities were not as
clearly evident among patients with high-normal BP stratified
by insulin resistance (Tables 2 and 3).
Neurohormonal Markers Plasma leptin, which is implicated in neurogenic hypertension,27 was higher in patients with the metabolic syndrome and with insulin
resistance than their respective controls, but the relationship to differences in heart rate was more variable. Resting
heart rate is a marker for neurogenic activation and a risk
factor for future hypertension and cardiovascular events.28
Resting heart rate was greater in insulin-resistant subjects
FIG. 2 The distribution is shown of body mass indices (lean ⬍25, overweight 25.0 to 29.9, and obese ⱖ30) among patients with the metabolic
syndrome (MS) and the two measures of insulin resistance (IR) (left). The percentages are shown of lean, overweight, and obese subjects
who had the metabolic syndrome and insulin resistance by HOMA and Matsuda indices (right).
8
Characteristic
1–2 MS risk
3–4 MS risk
HOMA IR(ⴚ)
HOMA IR(ⴙ)
Mats-DeF(ⴚ)
Mats-Def(ⴙ)
Number
Demographics
Age, years
Gender (M:F)
C:AA:other
Anthropometrics
Height (inches)
Weight (pounds)
BMI (kg/m2)
Waist circumference
(in.)
Waist:hip
Rest hemodynamics
SBP (mm Hg)
DBP (mm Hg)
HR (beats/min)
Metabolic risk factors
Fasting glucose (mg/dL)
Cholesterol (mg/dL)
LDL-C (mg/dL)
HDL-C (mg/dL)
Triglycerides (mg/dL)
TC/HDL
Neurohormones*
Leptin
Norepinephrine
Plasma renin
Angiotensin II
Aldosterone
88
53
80
48
75
44
49 ⫾ 1
67%
84%:15%:1%
48 ⫾ 1
70%
78%:20%:2%
50 ⫾ 1
64% / 34%
83% / 17%
47 ⫾ 1
73% / 27%
75% / 21% / 4%
49 ⫾ 1
67% / 33%
85%:15%
48 ⫾ 1
68% / 32%
70% / 28% / 2%
68.0 ⫾ 0.5
182.3 ⫾ 3.7
27.8 ⫾ 0.5
69.0 ⫾ 0.6
219.9 ⫾ 6.9
32.6 ⫾ 0.8
37.2 ⫾ 0.6
0.89 ⫾ 0.02
133 ⫾ 1
85 ⫾ 1
74 ⫾ 1
⫾
⫾
⫾
⫾
⫾
3
3
1
5
0.1
10.5
280
0.94
5.6
7.4
⫾
⫾
⫾
⫾
⫾
1.1
14
0.12
0.4
0.5
134 ⫾ 1
85 ⫾ 1
76 ⫾ 2
196
119
38
199
5.2
⫾
⫾
⫾
⫾
⫾
6
5
1
14
0.2
21.7
237
0.80
5.5
6.8
⫾
⫾
⫾
⫾
⫾
3.3
15
0.08
0.4
0.5
36.7 ⫾ 0.7
0.87 ⫾ 0.02
132 ⫾ 1
85 ⫾ 1
72 ⫾ 1
193
119
50
119
4.1
⫾
⫾
⫾
⫾
⫾
4
4
1
7
0.1
11.8
271
0.71
5.2
6.3
⫾
⫾
⫾
⫾
⫾
1.6
15
0.07
0.3
0.4
68.5 ⫾ 0.6
217.1 ⫾ 5.4
32.7 ⫾ 0.8
42.1 ⫾ 0.6
0.94 ⫾ 0.03
133 ⫾ 1
86 ⫾ 1
77 ⫾ 2
196
119
45
168
4.7
⫾
⫾
⫾
⫾
⫾
6
5
2
16
0.2
20.6
254
1.10
6.4
8.3
⫾
⫾
⫾
⫾
⫾
3.0
18
0.18
0.5
0.8
* Units for neurohormones same as Table 1 Data ⫽ mean ⫾ SEM. Significant P values and abbreviations are provided in Table 3.
68.6 ⫾ 0.5
186.1 ⫾ 4.3
27.9 ⫾ 0.6
37.0 ⫾ 0.7
0.87 ⫾ 0.02
133 ⫾ 1
85 ⫾ 1
72 ⫾ 1
193
118
50
120
4.1
⫾
⫾
⫾
⫾
⫾
4
4
2
8
0.1
12.1
268
0.78
5.2
6.8
⫾
⫾
⫾
⫾
⫾
1.8
14
0.07
0.3
0.5
67.8 ⫾ 0.7
211.5 ⫾ 5.6
32.3 ⫾ .7
41.6 ⫾ 0.6
0.94 ⫾ 0.03
133 ⫾ 1
85 ⫾ 1
77 ⫾ 2
196
124
45
158
4.6
⫾
⫾
⫾
⫾
⫾
6
6
2
16
0.2
21.5
257
1.11
6.5
8.0
⫾
⫾
⫾
⫾
⫾
3.1
21
0.2
0.6
0.7
AJH–January 2005–VOL. 18, NO. 1
192
120
53
98
3.8
41.2 ⫾ 0.7
0.92 ⫾ 0.03
68.3 ⫾ 0.5
182.2 ⫾ 3.7
27.5 ⫾ 0.5
METABOLIC SYNDROME AND INSULIN RESISTANCE
Table 2. Descriptive characteristics of TROPHY sub-study subjects stratified by metabolic syndrome (MS) and insulin resistance (IR) criteria
AJH–January 2005–VOL. 18, NO. 1
METABOLIC SYNDROME AND INSULIN RESISTANCE
9
Table 3. Comparison of statistically significant differences for key study variables based on stratification of
subjects by metabolic syndrome or insulin resistance criteria (data in Table 2)
Variable
Metabolic Syndrome
HOMA Index
Matsuda Index
****
****
****
****
****
****
*
***
****
****
†
**
**
**
**
***
**
†
***
**
****
*
*
**
**
****
***
***
****
****
****
*
†
**
****
****
****
****
****
**
**
**
†
†
*
†
Demographic and general health traits
Age
Gender (M:F)
Ethnicity, C:AA
Anthropometrics
Height (inches)
Weight (pounds)
BMI (kg/m2)
Waist circum (in)
Waist-hip ratio
Hemodynamics, rest
SBP (mm Hg)
DBP (mm Hg)
HR (beats/min)
Exercise
SBP, 3 min
SBP, 6 min
HR, 3 min
HR, 6 min
Metabolic variables
Chol (mg/dL)
LDL-C (mg/dL)
HDL-C (mg/dL)
Trig (mg/dL)
TC/HDL
Fasting glucose
1-h glucose
2-h glucose
Fasting insulin
1-h insulin
2-h insulin
Neurohormones
Leptin
Norepinephrine
Plasma renin activity
Angiotensin II
Aldosterone
*
****
****
****
**
*
*
Abbreviations used in Table 2 and Table 3. C ⫽ white; AA ⫽ African American; BMI ⫽ body mass index, kg/m2; S ⫽ systolic; D ⫽ diastolic;
BP ⫽ blood pressure; HR ⫽ heart rate; Chol ⫽ total cholesterol; LDL-C ⫽ low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein
cholesterol; trig ⫽ triglycerides; TC/HDL ⫽ total cholesterol to HDL ratio. Units for neurohormones are provided in Table 1.
† P ⬍ .05 by one-tail and ⬍.10 by two-tail test. * P ⬍ .05, ** P ⬍ .01, *** P ⬍ .001, **** P ⬍ .0001.
but not in metabolic syndrome patients relative to their
controls (Tables 2 and 3).
Previous reports suggest obesity and insulin resistance
are associated with a more active renin-angiotensin-aldosterone system,29 –31 which is implicated in cardiovascular
and renal pathology.32 Patients with insulin resistance
defined by HOMA and Matsuda-DeFronzo also tended to
have a more active renin-angiotensin-aldosterone system.
However, there was no evidence for greater activity of the
renin-angiotensin-aldosterone system among subjects with
the metabolic syndrome compared to their respective
controls.
Hemodynamic Responses Heart rates at 3 and 6 min of
treadmill exercise were higher among insulin-resistant patients than controls but were not different between patients
with and without the metabolic syndrome. The explanation
for these differences is not eludicated by the current study
but could reflect differences in sympathovagal responses to
physical stress.
Systolic BP after 6 min of standardized treadmill exercise was greater in patients with than without the metabolic syndrome and insulin resistance, compared to their
controls, despite virtually identical baseline BP values.
Systolic BP during exercise is predictive of cardiovascular
events independently of resting BP,33 indicating these
patients have yet another risk factor for cardiovascular
disease.
Limitations of our study include the absence of normal
volunteers, which impedes comparisons to previous reports
and extrapolation to individuals with normal or higher BPs.
Although plasma insulin was assayed in a single laboratory
for all volunteers in this study, the lack of a standardized
10
METABOLIC SYNDROME AND INSULIN RESISTANCE
AJH–January 2005–VOL. 18, NO. 1
FIG. 3 The glucose and insulin values are depicted under fasting conditions and at 1- and 2-h of a standard oral glucose tolerance test in
patients with high-normal blood pressure subdivided by the presence (filled circles) or absence (open circles) of metabolic syndrome, HOMA,
or Matsuda index insulin resistance. *P ⬍ .05; †P ⬍ .01: ‡ P ⬍ .001.
assay does not permit direct comparisons with previous reports. Therefore, insulin resistance was defined by the same
proportion of patients who met metabolic syndrome criteria,
that is, the upper ⬃3/8 (53 of 141 or 37.6%) of the distribution. Insulin resistance was defined using glucose and insulin
values rather than the gold standard euglycemic clamp. Nevertheless, these clinically applicable indices are strongly correlated with results from the glucose clamp.14,15
Other limitations of the study include plasma norepinephrine as a marker of sympathetic nervous system activity. Evidence suggests that the sympathetic nervous
system participates in obesity-related hypertension and
metabolic abnormalities and that hyperleptinemia and hyperinsulinemia can increase sympathetic activity. Differences in sympathetic activity between metabolic syndrome
or insulin-resistant groups and their respective controls
may have been identified if more sensitive measures such
as norepinephrine turnover or muscle sympathetic nerve
activity had been measured. In a similar vein, 24-h urine
aldosterone is likely a better index of aldosterone production than a single plasma aldosterone measurement.
Moreover, plasma renin activity and angiotensin II were
measured with volunteers supine, which would be associated with lower circulating levels that could tend to obscure differences. The lack of a purification step before the
angiotensin II assay would further obscure differences
between groups. Thus, the failure to detect significant
differences in renin-angiotensin-aldosterone system activ-
ity between subjects with and without the metabolic syndrome may reflect the limitations noted. Obesity and body
fat distribution were assessed by body mass index, waist
circumference and waist/hip ratios. Dual-beam x-ray absorptiometry and computed tomographic scans would
have provided more precise estimates of adiposity and
body fat distribution and may have yielded different or
stronger associations with other variables measured in this
study. Finally, volunteers were predominantly white,
which limits extrapolation to other ethnic groups including
African Americans.
Summary
The prevalence of the metabolic syndrome is ⬃50% higher
among adults with high-normal BP (⬃3/8) than the general
population (⬃1/4). Metabolic syndrome and insulin resistance are both strongly related to body mass index, but they
are not synonymous. Waist circumference, fasting triglycerides, glucose, leptin, and systolic BP after 6 min of exercise
were consistently different for both the metabolic syndrome
and insulin resistance designations compared to their respective controls. The groups showed less consistent differences
compared to controls for waist/hip ratio, HDL-cholesterol,
heart rate responses to exercise, and markers of renin-angiotensin-aldosterone system activity. These differences may
have implications for the pathogenesis of cardiovascular risk
and strategies of prevention. The observations point to the
AJH–January 2005–VOL. 18, NO. 1
METABOLIC SYNDROME AND INSULIN RESISTANCE
11
FIG. 4 The systolic blood pressure (SBP) and heart rate (HR) values are shown under resting baseline conditions (standing) and at 3 and 6
min of a modified Bruce protocol among patients with high-normal blood pressure subdivided by the presence (filled circles) or absence (open
circles) of the metabolic syndrome, HOMA, or Matsuda index insulin resistance. *P ⬍ .05; †P ⬍ .01; ‡P ⬍ .001.
need for further refinement of definitions that will facilitate
translation of research findings into practice for the growing
ranks of metabolic syndrome/insulin resistant patients. In the
interim, preventing and treating overweight and obesity remain vital in efforts to reduce the growing burden of metabolic syndrome and insulin resistance-related disease.8,34
iew; Cleveland Clinic Foundation: Donald Vidt, MD, Fetnat
Fouad-Tarazi, MD, Rita Spirko, Marykay Dapaul.
Acknowledgments
2.
TROPHY Sub-study Investigative Team: The data represent
the dedicated effort of many individuals including: Astra
Zeneca: Melissa Grozinski, Senior Clinical Research Scientist; Eric Michelson, MD, Senior Medical Director. University of Michigan: Stevo Julius, MD, ScD, David Freido, Liz
O’Connor; Investigators and study coordinators at Sub-study
sites: University of Alabama: David Calhoun, MD, Karen
Scaglioni, Sharon Green; University of Minnesota: Jay Cohn,
MD, Natalia Florea, Liza Bordeaux; University of Tennesse
(Memphis, VA): William Cushman, MD, Carmen DeJesus,
Cathy Thompson; Medical University of South Carolina:
Brent Egan, MD, Linda Ambrose, Jackie Nguyen; The Western Pennsylvania Hospital: Alan Gradman, MD, Monica
Reda; University of Mississippi; Marion Wofford, MD,
Kathy Adair; Orange County Research Center: Joel Neutel,
MD, Rebecca Lopez; Georgetown University Medical Center (Department of Memphis Veterans Affairs Hospital and
Clinic): Vasilios Papademetriou, MD, Susan Borgiaz; Medical College of Virginia: Domenic Sica, MD, Rhonda Trim-
References
1.
3.
4.
5.
6.
7.
8.
Prebble WE: Obesity: Observations on one thousand cases. Boston
Med Surg J 1923;88:617– 621.
World Health Organization: Obesity. Preventing and managing the
global epidemic: report of a WHO Consultation on Obesity, Geneva,
3–5 June 1997, World Health Organization, 1998.
Flegal KM, Carroll MD, Ogden CL, Johnson CL: Prevalence and
trends in obesity among US adults, 1999 –2000. JAMA 2002;288:
1723–1727.
Kissebah AH, Krakower GR: Regional adiposity and morbidity.
Physiol Rev 1994;74:761– 811.
Reaven GM: Pathophysiology of insulin resistance in human disease. Physiol Rev 1995;75:473– 486.
Ferrannini E, Natali A, Bell P, Cavallo-Perin P, Lalic N, Mingrone
G: Insulin resistance and hypersecretion in obesity. J Clin Invest
1997;100:1166 –1173.
National Cholesterol Education Program Expert Panel on Detection,
Evaluation, and Treatment of High Blood Cholesterol in Adults:
Executive Summary of the Third Report of the National Cholesterol
Education Program (NCEP) Expert Panel on Detection, Evaluation,
and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA 2001;285:2486 –2497.
Park Y-W, Zhu S, Palaniappan L, Heshka S, Carnethon MR, Hemsfield
SB: The metabolic syndrome: prevalence and associated risk factor
findings in the US population from the Third National Health and
Nutrition Examination Survey, 1988 –1994. Arch Intern Med 2003;
163:427– 436.
12
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
METABOLIC SYNDROME AND INSULIN RESISTANCE
AJH–January 2005–VOL. 18, NO. 1
Robinson SC, Brucer M: Range of normal blood pressure: a statistical and clinical study of 11, 383 persons. Arch Int Med 1939;64:
409 – 443.
Rabkin SW, Mathewson FAL, Tate RB: Relationship of blood
pressure in 20 –39-year-old men to subsequent blood pressure and
incidence of hypertension over a 30-year observation period. Circulation 1982;65:291–299.
Kazumi T, Kawaguchi A, Sakai K, Hirano T, Yoshino G: Young
men with high-normal blood pressure have lower serum adiponectin, smaller LDL size, and higher elevated heart rate than those with
optimal blood pressure. Diabetes Care 2002;25:971–976.
Nesbitt SD, Julius S: Prehypertension: a possible target for antihypertensive medication. Curr Hypertens Rep 2000;2:356 –361.
Nesbitt SD, Egan BM, Julius S: High normal blood pressure and
co-morbidity in the TROPHY Study. Am J Hypertension 2002;15:
208A.
Matsuda M, DeFronzo RA: Insulin sensitivity indices obtained from
oral glucose testing: Comparison with the euglycemic insulin clamp.
Diabetes Care 1999;22:1462–1470.
Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF,
Turner RC: Homeostasis model assessment: insulin resistance and
beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412– 419.
Joint National Committee on Prevention, Detection, Evaluation, and
Treatment of High Blood Pressure: The sixth report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment
of High Blood Pressure. Arch Intern Med 1997;157:2413–2446.
McVeigh GE, Bratteli CW, Morgan DJ, Alinder CM, Glasser SP,
Finkelstein SM, Cohn JN: Age-related abnormalities in arterial
compliance identified by pressure pulse contour analysis. Hypertension 1999;33:1392–1398.
Krahn AD, Klein GJ, Yee R: Hysteresis of the RT interval with
exercise: a new marker for the long-QT syndrome? Circulation
1997;96:1551–1556.
Expert Committee on the Diagnosis and Classification of Diabetes
Mellitus: Report of the Expert Committee on the Diagnosis and
Classification of Diabetes Mellitus. Diabetes Care 1999;22:S5–S19.
Emanuel RL, Cain JP, Williams GH: Double antibody radioimmunoassay of renin activity and angiotensin II in human peripheral
plasma. J Lab Clin Med 1973;81:632– 640.
Lakka H-M, Laaksonen DE, Lakka TA, Niskanen LK, Kumpusalo
E, Tuomilehto J, Salonen JT: The metabolic syndrome and total and
cardiovascular disease mortality in middle-aged men. JAMA 2002;
288:2790 –2716.
Laaksonen DE, Lakka HM, Niskanen LK, Kaplan GA, Salonen JT,
Lakka TA: Metabolic syndrome and development of diabetes mel-
litus: application and validation of recently suggested definitions of
the metabolic syndrome in a prospective cohort study. Am J
Epidemiol 2002;156:1070 –1077.
Stern M, Haffner S: Body fat distribution and hyperinsulinemia as
risk factors for diabetes and cardiovascular disease. Arteriosclerosis
1986;6:123–129.
Kannel WB, Cupples LA, Ramaswami R, Stroke J, Kreger BE,
Higgins M: Regional adiposity and risk of cardiovascular disease:
the Framingham Study. J Clin Epidemiol 1991;44:183–190.
The European Group for the Study of Insulin Resistance (EGIR):
Frequency of the WHO metabolic syndrome in European cohorts,
and an alternative definition of an insulin resistance syndrome.
Diabetes Metab 2002;28:364 –376.
Solymoss BC, Bourassa MG, Campeau L, Sniderman A, Marcil M,
Lespérance J, Lévesque S, Varga S: Effect of increasing metabolic
syndrome score on atherosclerotic risk profile and coronary artery
disease angiographic severity. Am J Cardiol 2004;93:159 –164.
Mark AL, Correia M, Morgan DA, Shaffer RA, Haynes WG:
State-of-the-art lecture: obesity-induced hypertension: new concepts
from the emerging biology of obesity. Hypertension 1999;33:537–
541.
Palatini P, Edoardo C, Pauletto P, Staessen J, Kaciroti N, Julius S:
Relationship of tachycardia with high normal blood pressure and
metabolic abnormalities: a study with mixture analysis in three
populations. Hypertension 1997;30:1267–1273.
Sowers JR, Nyby M, Stern N, Beck F, Baron S, Catania R, Vlachis
N: Blood pressure and hormone changes associated with weight
reduction in the obese. Hypertension 1982;4:686 – 691.
Rocchini AP, Katch VL, Grekin R, Morehead C, Anderson J: Role
of aldosterone in blood pressure regulation in obese adolescents.
Am J Cardiol 1986;57:613– 618.
Egan BM, Stepniakowski K, Goodfriend TL: Renin and aldosterone
are higher and the hyperinsulinemic effects of salt restriction greater
in subjects with risk factor clustering. Am J Hypertension 1994;7:
886 – 893.
Heart Outcomes Prevention Evaluation Study Investigators: Effects
of ramipril on cardiovascular and microvascular outcomes in people
with diabetes mellitus: results of the HOPE study and MICROHOPE substudy. Lancet 2000;355:253–259.
Mundal R, Kjeldsen SE, Sandvik L, Erikssen G, Thaulow E, Erikssen
J: Exercise blood pressure predicts mortality from myocardial infarction. Hypertension 1996;27:324 –329.
Case CC, Jones PH, Nelson K, O’Brian Smith E, Ballantyne CM:
Impact of weight loss on the metabolic syndrome. Diab Obes Metab
2002;4:407– 414.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.