OBJECTIVE—To determine the meaning of Si = 0 derived from the frequently sampled intravenous glucose tolerance test.

RESEARCH DESIGN AND METHODS—The issue of assessing insulin resistance in large studies is important because the most definitive method (“gold standard”), the hyperinsulinemic-euglycemic clamp, is expensive and invasive. The frequently sampled intravenous glucose tolerance test (FSIGTT) has been widely used, but in insulin-resistant subjects (especially diabetic subjects), it yields considerable numbers of subjects whose Si is zero. The interpretation of an Si equaling zero is unknown.

RESULTS—To address this issue, we examined 1,482 subjects from the Insulin Resistance Atherosclerosis Study (IRAS) using an insulin-modified FSIGTT and minimal model calculation of Si. The proportion of insulin-resistant subjects (Si < 1.61 × 10−4 [min−1 · μU−1 · ml−1] based on the median of the nondiabetic population) was 38.6% in subjects with normal glucose tolerance (NGT), 74% in subjects with impaired glucose tolerance (IGT), and 92% in subjects with type 2 diabetes. The proportion of subjects with Si = 0 was 2.2% in subjects with NGT, 13.2% in subjects with IGT, and 35.7% in subjects with type 2 diabetes. In subjects with IGT, those with Si = 0 had significantly lower HDL cholesterol levels and higher BMI, waist circumference, fibrinogen, plasminogen-activator inhibitor 1 (PAI-1), C-reactive protein (CRP), and 2-h insulin levels than insulin-resistant subjects with Si > 0. In type 2 diabetes, subjects with Si = 0 had significantly greater BMI and waist circumference and higher triglyceride, PAI-1, CRP, fibrinogen, and fasting and 2-h insulin levels than insulin-resistant subjects with Si > 0. In addition, diabetic subjects with Si = 0 had more metabolic disorders related to the insulin resistance syndrome than diabetic insulin-resistant subjects with Si > 0.

CONCLUSIONS—We found very few subjects with Si = 0 among subjects with NGT and few subjects with Si = 0 among subjects with IGT. In contrast, Si = 0 was common in subjects with diabetes. Subjects with Si = 0 tended to have more features of the insulin resistance syndrome than other insulin-resistant subjects with Si > 0, as would be expected of subjects with almost no insulin-mediated glucose disposal, thus suggesting that subjects with Si = 0 are correctly classified as being very insulin resistant rather than having failed the minimal model program.

Hyperinsulinemia and insulin resistance have been related to the development of type 2 diabetes (17) and cross-sectionally and prospectively with cardiovascular risk factors and atherosclerosis (817). Most studies (especially large population-based studies) use surrogates for insulin resistance such as fasting insulin (18) because of the expense and difficulty of direct measures of insulin resistance. In populations in which both insulin levels and insulin resistance have been measured, the latter often has been more closely associated with important clinical outcomes. For example, insulin resistance (as determined by the hyperinsulinemic-euglycemic clamp) was more closely correlated with the development of type 2 diabetes in Pima Indians than was fasting insulin concentration (7). Similarly, insulin resistance (determined by the frequently sampled intravenous glucose tolerance test [FSIGTT]) with minimal model was more closely correlated with atherosclerosis as determined by carotid wall thickness than were insulin concentrations per se in the Insulin Resistance Atherosclerosis Study (IRAS) (17).

The hyperinsulinemic-euglycemic clamp (19), which is the most widely accepted method to assess insulin resistance, is expensive and labor intensive. The FSIGTT has also been used to assess insulin resistance (20,21). A number of modifications have been used to increase the generalizability of the FSIGTT by the use of insulin injections in diabetic subjects (22) and reducing the number of blood samples required (n = 12) (23). Nevertheless, the use of this technique resulted in a number of subjects whose calculated Si = 0 by the minimal model computer program in more insulin-resistant subjects, especially diabetic subjects. In a small group of subjects (n = 55), Saad et al. (24) described the prevalence of Si = 0 (type 2 diabetes: 50% [12/24]; impaired glucose tolerance [IGT]: 15% [3/20]; normal glucose tolerance [NGT]: 0% [0/11]) using an insulin-modified protocol with 12 time points. A number of explanations for the Si = 0 are possible. The first is that these subjects are, indeed, very insulin resistant with insulin sensitivity not distinguishable from zero. A second possibility is that the use of a one-compartment model (25) may underestimate the Si, although this interpretation was not supported in other studies (26). Another possibility is that the FSIGTT may yield lower estimates of glucose disposal than the clamp because of the use of a short-acting bolus with its consequent high peak of insulin (27) rather than hyperinsulinemia of long duration, as with the clamp (28).

We have shown that cardiovascular risk factors are increased in insulin-resistant diabetic subjects relative to insulin-sensitive diabetic subjects (29). In this report, we examined whether insulin-resistant diabetic subjects with Si = 0 have increased metabolic syndrome risk factors relative to insulin-resistant subjects with Si > 0. To examine this issue, we first elucidated the frequency of Si = 0 in the IRAS, a population-based study of cardiovascular risk factors and insulin sensitivity (30). Next, we characterized all the subjects as insulin resistant or insulin sensitive by using the median for Si in the nondiabetic population, as had been done previously (31,32). We then evaluated whether subjects with Si = 0 had more features associated with the insulin resistance syndrome (hyperinsulinemia, obesity, upper-body adiposity, increased dyslipidemia, hypertension, and impaired fibrinolysis and enhanced coagulation) than other insulin-resistant subjects (subjects with Si > 0 but less than the median for Si in nondiabetic subjects). These analyses are presented separately by glucose tolerance status.

A detailed description of the design and methods of the IRAS has been published (30). In brief, this study was conducted at four clinical centers: Oakland and Los Angeles, California; San Antonio, Texas; and San Luis Valley, Colorado. Diabetic subjects on insulin were not eligible for the IRAS. Of all eligible subjects contacted, 48% completed the 2-day IRAS examination. Diabetic subjects with a fasting glucose level ≥300 mg/dl (≥16.7 mmol/l) were excluded.

A total of 1,625 individuals participated in the IRAS (56% women) (30). Individuals with NGT comprised the largest segment of the study sample (44%) (non-Hispanic white, n = 291; African American, n = 187; and Hispanic, n = 241), followed by those with diabetes (33%) (non-Hispanic white, n = 177; African American, n = 176; and Hispanic, n = 241) and those with IGT (23%) (non-Hispanic white, n = 145; African American, n = 101; and Hispanic, n = 123). The distribution of insulin sensitivity has been recently described in nondiabetic subjects (31) and diabetic subjects (32) from the IRAS.

Height, weight, and girths (minimum waist, waist at the umbilicus and hips) were measured following a standardized protocol. BMI (weight/height2 [kg/m2]) was used as an estimate of overall adiposity. Waist circumference was taken as the minimum circumference between the thorax and the hips. The waist circumference was used as an estimate of body fat distribution (30).

The IRAS examination required two visits (∼1 week apart [range 2–28 days]) (3032), each lasting ∼4 h. An oral glucose tolerance test and FSIGTT were performed during the first and second visits, respectively. Glucose tolerance was classified according to the World Health Organization criteria (33).

Resting systolic blood pressure and fifth-phase blood pressure were measured three times, and the second and third measurements were averaged. Hypertension was defined as systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥90 mmHg or current use of antihypertensive medication.

Insulin resistance was assessed by the FSIGTT (20) with minimal model analyses (34). Two modifications of the original protocol were used. An injection of insulin, rather than tolbutamide, was used to ensure adequate plasma insulin levels for the accurate computation of insulin resistance across a broad range of glucose tolerance (22). This was necessary because of the blunted or absent insulin response in diabetic subjects. Also, the reduced sampling protocol (which required 12 rather than 30 plasma samples and shows similar results to the full protocol [23]) was used because of the large number of subjects. Glucose in the form of a 50% solution (0.3 g/kg) and regular human insulin (0.03 units/kg) were injected through an intravenous line at 0 and 20 min, respectively. Blood was collected at −5, 2, 4, 8, 19, 22, 30, 40, 50, 70, 100, and 180 min for plasma glucose and insulin concentrations. Si was calculated by mathematical modeling methods using the MINMOD program (version 3.0 [1994]). This modified version of the FSIGTT protocol used in the IRAS has been compared with the hyperinsulinemic-euglycemic clamp (24). Acute insulin response was calculated as the increase in insulin concentrations at 2–8 min above the basal (fasting) insulin level.

Plasma glucose was measured with the glucose oxidase technique on an automated autoanalyzer (Yellow Springs Instruments). Insulin was measured using the dextran-charcoal radioimmunoassay, which has considerable cross-reactivity with proinsulin.

Plasma lipoprotein measurements were obtained from fasting single fresh plasma samples using the Lipid Research Clinic methods. VLDL was isolated by preparative ultracentrifugation, and VLDL (top) and bottom fractions were measured for cholesterol and triglyceride concentrations. HDL cholesterol was measured after precipitation of apolipoprotein B-containing lipoproteins with MnCl2 and heparin. The cholesterol content in the supernatant was measured in a separate autoanalyzer channel set to measure low cholesterol values. LDL cholesterol was calculated as the difference between the HDL cholesterol and the directly measured VLDL bottom cholesterol. Triglycerides were measured enzymatically after correction for free glycerol.

LDL size distribution (i.e., distribution of diameter of the major LDL peak for each participant) was determined using the method of Krauss and Burke (35). Gradient gels were obtained from Isolab (Akron, OH). Measurement of the size of the predominant peak was calibrated using LDL subfractions, the molecular diameter of which was determined by analytical ultracentrifugation (courtesy of Dr. R. Krauss, Donner Laboratories, Berkeley, CA). The LDL size of the predominant peak for an individual was defined as that person’s LDL size (36).

Fibrinogen was measured in citrated plasma with a modified clot-rate assay using the Diagnostica STAGO ST4 instrument, as described elsewhere (37). This was based on the original method of Clauss (38) with an internal coefficient of variation (CV) of 3.0%. Plasminogen activator inhibitor 1 (PAI-1) was also measured in citrated plasma (39), using a two-site immunoassay that is sensitive to free PAI-1 but not to PAI-1 complex with tissue plasminogen activator (t-PA) (40); the internal CV was 6.0%. The citrate sample was centrifuged for a minimum of 30,000g per minute to make certain that there was no contamination from platelet PAI-1. C-reactive protein (CRP) was measured using an in-house ultrasensitive competitive immunoassay (antibodies and antigens from Calbiochem, La Jolla, CA) with an interassay CV of 8.9% (41).

Mean values of the cardiovascular risk factors were compared according to insulin sensitivity by ANCOVA (SAS version 6.08; SAS Institute, Cary, NC). Logarithmic transformations (for statistical testing) were used for triglyceride, VLDL cholesterol, VLDL triglyceride, and PAI-1. Further adjustment was made for waist circumference. Because waist circumference and BMI were highly correlated (r = 0.82), they were not included in the same regression model. Adjustment for waist-to-hip ratio rather than for waist circumference yielded similar results. We preferred to present data for waist circumference rather than waist-to-hip ratio to provide a better measure of visceral adiposity (42). We initially presented our data separately by ethnic group. Using multiple linear regression, we tested for the interaction of Si = 0 by ethnicity in insulin-resistant subjects. We found no evidence of significant interactions, suggesting that the effect of Si = 0 on cardiovascular risk factors and adiposity was similar in each ethnic group. We also tested for the interaction of sex × Si = 0; again, these interactions were not significant. Therefore, we present data pooling the ethnic groups and both sexes. P values for dichotomous or categorical variables were calculated by the χ2 test. The prevalence of the National Cholesterol Education Program (NCEP) (43) definition of the metabolic syndrome in relation to Si = 0 or Si > 0 in subjects with IGT or diabetes was calculated (Fig. 1). P values were calculated by χ2.

Table 1 shows the distribution of Si by glucose tolerance status. The mean Si (×10−4 [min−1 · μU−1 · ml−1]) was 2.62 ± 0.32 in subjects with NGT, 1.26 ± 0.52 in subjects with IGT, and 0.55 ± 0.01 in subjects with type 2 diabetes (P < 0.001). The proportion of subjects who were insulin resistant (Si <1.61, median for Si in nondiabetic subjects) was 38.6% in subjects with NGT, 74.0% in subjects with IGT, and 92.0% in subjects with type 2 diabetes. The number of subjects with Si = 0 was 2.2% in subjects with NGT, 13.2% in subjects with IGT, and 35.7% in subjects with type 2 diabetes. Because few subjects with NGT had Si = 0, subjects with NGT will not be considered further in this article. The remainder of this article will consider insulin-resistant subjects with IGT (n = 246) and type 2 diabetes (n = 442). We will consider whether subjects with Si = 0 are different from subjects with Si > 0 in terms of variables related to the metabolic syndrome.

Table 2 shows levels of anthropometric and cardiovascular risk factors among insulin-resistant type 2 diabetic subjects according to whether they have Si = 0 or Si > 0 adjusted for age, sex, ethnicity, and clinic. Subjects with Si = 0 had significantly greater BMI, waist circumference, triglyceride, PAI-1, fibrinogen, CRP, and fasting and 2-h insulin levels than subjects with Si > 0. Table 3 shows similar data after further adjustment for waist circumference. Subjects with Si = 0 continued to have significantly greater PAI-1, CRP, and fasting and 2-h insulin levels (Fig. 1) than subjects with Si > 0, although the differences were considerably attenuated.

Table 4 shows the levels of anthropometric and cardiovascular risk factors among insulin-resistant IGT subjects according to whether they had Si = 0 or Si > 0. Subjects with Si = 0 had significantly higher BMI and waist circumference, 2-h insulin, fibrinogen, and CRP levels and lower HDL cholesterol levels than subjects with Si > 0. After further adjustment for waist circumference, subjects with Si = 0 continued to have significantly greater 2-h insulin, CRP, and fibrinogen levels and lower HDL cholesterol levels than subjects with Si > 0 (Table 5).

Figure 1 shows an analysis of clustering of variables related to the metabolic syndrome according to whether subjects had Si = 0 or Si > 0. Five factors were identified: 1) high triglyceride, 2) upper-body adiposity (high waist circumference), 3) fasting ≥110 mg/dl, 4) low HDL cholesterol, and 5) hypertension. The cut points were based on the NCEP criteria for the metabolic syndrome (45). Individuals could have zero to four disorders. In both subjects with IGT and subjects with type 2 diabetes, those with Si = 0 had a shift to more metabolic disorders than those with Si > 0, although these results were significant only for the type 2 diabetic subjects. The prevalence of the NCEP metabolic syndrome in IGT subjects was 51.2% in subjects with Si = 0 compared with 46.6% in subjects with Si: 0 < Si < 1.61 (NS) (Fig. 2). The prevalence of the NCEP metabolic syndrome in diabetic subjects with Si = 0 was 85.2% compared with 70.8% in subjects with Si: 0 < Si < 1.61 (P < 0.001).

Among a group of subjects with insulin resistance (defined by Si < 1.61 × 10−4 [min−1 · μU−1 · ml−1] based on the median in the nondiabetic population), we have shown that subjects with IGT and type 2 diabetes with Si = 0 are significantly more obese (as determined by BMI) and have greater upper-body adiposity (as determined by waist circumference) than subjects with 0 < Si < 1.61. (This was also true of subjects with NGT, although the number of subjects with Si = 0 was very small [n = 15] and therefore not shown in the tables.) We have also shown that subjects with Si = 0 have increased cardiovascular risk factors compared with subjects with Si > 0, although the results were not completely consistent in the IGT and type 2 diabetic subjects (lipids: type 2 diabetes [increased triglyceride] vs. IGT [decreased HDL cholesterol]; fibrinolysis/coagulation: type 2 diabetes [increased PAI-1 and fibrinogen and subclinical inflammation, increased CRP in both type 2 diabetes and IGT] vs. IGT [increased fibrinogen]). Blood pressure did not differ in insulin-resistant subjects with Si = 0 vs. Si > 0. Lastly, subjects with Si = 0 had higher fasting and 2-h insulin concentrations than subjects with Si > 0, in both IGT and type 2 diabetes. The differences between subjects with Si = 0 and Si > 0 were only partially associated with the increased upper-body adiposity in subjects with Si = 0 (Tables 3 and 5). Additionally, subjects with Si = 0 had higher insulin concentrations after further adjustments for the small differences in the glucose concentrations between Si = 0 and Si > 0 subjects (data not shown). Taken together, these findings indicate that subjects with Si values indistinguishable from zero were more insulin resistant than their insulin-resistant counterparts with Si > 0. These results are reinforced by the evidence of greater clustering of cardiovascular risk factors in diabetic subjects with Si = 0 than in subjects with Si > 0, and a higher prevalence of the metabolic syndrome defined by the NCEP (Fig. 1). These results were significant in diabetic subjects but not in subjects with IGT possibly because of the much lower number of IGT subjects with Si = 0 than diabetic subjects with Si = 0 (n = 44 vs. 172) (Fig. 2).

Because laboratory procedures such as the glucose clamp are not practical in a large study, we used the minimal model. Strong correlations between Si from the minimal model and glucose disposal rate from the clamp have been reported in several studies (24). In normal subjects, interpretable measurements of Si were derived from the insulin-boosted FSIGTT. However, in IRAS, we discovered in some IGT subjects (13.2%) and in many participants with type 2 diabetes (35.7%) that it was not possible to calculate a value of Si from the MINMOD software that was distinguishable from 0. The purpose of the present analysis was to examine characteristics of subjects with Si not distinguishable from 0; we could note these values of Si as “Si ∼ 0” for ease of discussion.

Our data lend support to the notion that zero Si values obtained from minimal model analysis of the insulin-modified FSIGTT represent a lack of a discernable effect of the injected amount of insulin on plasma glucose. To clarify this issue further, it is necessary to recapitulate the approach used to estimate Si with the minimal model approach. The MINMOD program examines the moment-to-moment effect of the changes in insulinemia on plasma glucose and calculates a value for Si. The insulin sensitivity index obtained with this approach is simply the steady-state effect of an incremental change in plasma insulin to increase fractional glucose disappearance independent of glycemia. In extremely insulin-resistant subjects, the injected amount of insulin (∼2 units in the current study) fails to produce a discernable change in glucose utilization. Consequently, the model cannot assign a finite value to Si and a zero value is obtained.

Therefore, the MINMOD Si = 0 values appear to identify a group of subjects (mostly type 2 diabetic patients) in whom insulin-mediated glucose disposal is very low. The existence of such very insulin-resistant subjects is supported by DeFronzo et al. (19), who showed that insulin infusion during the clamp at a rate of 40 mU · m−2 · min−1 (a total dose of ∼16 units over 3 h) increased plasma insulin concentrations to 531 ± 102 pmol/l without inducing a significant increase in forearm glucose uptake (5.84 ± 1.51 μmol · min−1 · kg−1 vs. a basal value of 4.38 ± 1.16). Moreover, Alzaid et al. (44) found that when the plasma insulin pattern normally seen during an oral glucose tolerance test was simulated by an intravenous insulin infusion, while clamping glucose at the basal concentration, the insulin increment had no measurable effect on the glucose utilization rate in type 2 diabetic patients. The total insulin dose infused in the latter study was similar to that used in the insulin-modified FSIGTT with an insulin dose of 0.03 units/kg, viz., ∼2 units. These findings suggest that MINMOD Si = 0 values represent a real pathophysiological phenomenon (i.e., a lack of glucose response to an increase in insulin level within the range that occurs with day-to-day food ingestion) that exists in a substantial proportion of some individuals with type 2 diabetes and in a minority of those with IGT or NGT.

The current report of the IRAS on whether Si = 0 subjects are insulin resistant was a retrospective analysis. A more definitive approach to whether Si = 0 subjects are actually very insulin resistant would be to use prospectively the euglycemic-hyperinsulinemic clamps in subjects whose FSIGTT showed an Si = 0. The issue of Si = 0 is most important in diabetic subjects because of the higher prevalence of Si = 0 in this group.

In conclusion, we have shown that subjects with IGT and type 2 diabetes who have Si = 0 are more obese and have increased cardiovascular risk factors linked to the insulin resistance syndrome and greater peripheral hyperinsulinemia than corresponding insulin-resistant subjects with Si > 0. These results suggest that subjects with Si = 0 are, indeed, very insulin resistant and probably represent an Si very close to zero rather than a failure of the minimal model. Perhaps these subjects might be better described as having insulin sensitivity not distinguishable from 0 (Si ∼ 0).

Figure 1—

Mean insulin levels and standard errors adjusted for age, sex, and ethnicity in IGT and diabetes. ▪, Si = 0; □, 0 < Si < 1.61.

Figure 1—

Mean insulin levels and standard errors adjusted for age, sex, and ethnicity in IGT and diabetes. ▪, Si = 0; □, 0 < Si < 1.61.

Close modal
Figure 2—

Relation of numbers of metabolic disorders (0–5) in relation to Si = 0 or Si > 0 in insulin-resistant subjects. A: All subjects. B: Diabetic subjects. P values were calculated by χ2. Metabolic disorders were defined by the NCEP criteria (43). □, Si = 0 (n = 231); ▪, 0 < Si < 1.61 (n = 716); P < 0.0001.

Figure 2—

Relation of numbers of metabolic disorders (0–5) in relation to Si = 0 or Si > 0 in insulin-resistant subjects. A: All subjects. B: Diabetic subjects. P values were calculated by χ2. Metabolic disorders were defined by the NCEP criteria (43). □, Si = 0 (n = 231); ▪, 0 < Si < 1.61 (n = 716); P < 0.0001.

Close modal
Table 1—

Distribution of clinical characteristics of subjects by glucose tolerance status (including both insulin-resistant and insulin-sensitive subjects)

NGTIGTType 2 diabetes
n 671 332 479
Si (×10–4 [min−1 · μU−1 · ml−1]) 2.62 ± 0.41 1.26 ± 0.52 0.55 ± 0.32
Si = 0 15 (2.2) 44 (13.2) 172 (35.7)
Si < 1.61 (insulin resistant) 259 (38.6) 246 (74.0) 442 (92.0)
NGTIGTType 2 diabetes
n 671 332 479
Si (×10–4 [min−1 · μU−1 · ml−1]) 2.62 ± 0.41 1.26 ± 0.52 0.55 ± 0.32
Si = 0 15 (2.2) 44 (13.2) 172 (35.7)
Si < 1.61 (insulin resistant) 259 (38.6) 246 (74.0) 442 (92.0)

Data are n, means ± SD, or n (%).

Table 2—

Clinical characteristics of insulin-resistant diabetic subjects by Si = 0 or Si > 0 (0 < Si < 1.61) adjusted for age, sex, ethnicity, and clinic

Si = 00 < Si < 1.61P
n 172 270
Age (years)* 56.6 ± 0.7 57.2 ± 0.5 0.51
Ethnicity* (% W/AA/H) 36/30/34 31/34/35 0.60
Sex* (% female) 59 52 0.13
BMI (kg/m232.7 ± 0.4 31.0 ± 0.3 0.004
Waist circumference (cm) 102.5 ± 0.9 98.2 ± 0.7 <0.001
Waist-to-hip ratio (cm) 0.92 ± 0.01 0.91 ± 0.01 0.11
Cholesterol (mg/dl)
Total 212.1 ± 3.5 215.9 ± 2.5 0.37
HDL 39.1 ± 0.8 40.8 ± 0.7 0.12
LDL 138.7 ± 2.7 143.8 ± 35.6 0.17
Triglyceride (mg/dl) 210.0 ± 16.5 177.7 ± 6.7 0.039
LDL size (Å) 257.5 ± 0.6 265.4 ± 0.8 0.25
PAI-1 (ng/ml) 38.5 ± 2.2 29.8 ± 1.2 <0.001
Fibrinogen (mg/dl) 307.5 ± 5.0 288.2 ± 3.6 0.002
CRP (mg/l) 3.86 ± 0.30 2.83 ± 0.17 0.001
Glucose (mg/dl)
Fasting 172.1 ± 4.2 175.4 ± 3.8 0.51
2-h 313.2 ± 7.0 315.3 ± 5.7 0.87
Insulin (μU/ml)
Fasting 29.7 ± 1.5 21.0 ± 0.7 <0.001
2-h 120.7 ± 8.1 91.0 ± 4.8 <0.001
Si (×10−4 [min−1 · μU−1 · ml−1]) 0.00 0.62 ± 0.02 <0.001
Acute insulin response (μU · ml−1 · min−17.45 ± 2.5 6.85 ± 1.1 0.74
Blood pressure (mmHg)
Systolic 127.3 ± 1.15 126.9 ± 0.91 0.81
Diastolic 78.4 ± 0.69 78.1 ± 0.54 0.72
Hypertension prevalence* (%) 57 50 0.14
Si = 00 < Si < 1.61P
n 172 270
Age (years)* 56.6 ± 0.7 57.2 ± 0.5 0.51
Ethnicity* (% W/AA/H) 36/30/34 31/34/35 0.60
Sex* (% female) 59 52 0.13
BMI (kg/m232.7 ± 0.4 31.0 ± 0.3 0.004
Waist circumference (cm) 102.5 ± 0.9 98.2 ± 0.7 <0.001
Waist-to-hip ratio (cm) 0.92 ± 0.01 0.91 ± 0.01 0.11
Cholesterol (mg/dl)
Total 212.1 ± 3.5 215.9 ± 2.5 0.37
HDL 39.1 ± 0.8 40.8 ± 0.7 0.12
LDL 138.7 ± 2.7 143.8 ± 35.6 0.17
Triglyceride (mg/dl) 210.0 ± 16.5 177.7 ± 6.7 0.039
LDL size (Å) 257.5 ± 0.6 265.4 ± 0.8 0.25
PAI-1 (ng/ml) 38.5 ± 2.2 29.8 ± 1.2 <0.001
Fibrinogen (mg/dl) 307.5 ± 5.0 288.2 ± 3.6 0.002
CRP (mg/l) 3.86 ± 0.30 2.83 ± 0.17 0.001
Glucose (mg/dl)
Fasting 172.1 ± 4.2 175.4 ± 3.8 0.51
2-h 313.2 ± 7.0 315.3 ± 5.7 0.87
Insulin (μU/ml)
Fasting 29.7 ± 1.5 21.0 ± 0.7 <0.001
2-h 120.7 ± 8.1 91.0 ± 4.8 <0.001
Si (×10−4 [min−1 · μU−1 · ml−1]) 0.00 0.62 ± 0.02 <0.001
Acute insulin response (μU · ml−1 · min−17.45 ± 2.5 6.85 ± 1.1 0.74
Blood pressure (mmHg)
Systolic 127.3 ± 1.15 126.9 ± 0.91 0.81
Diastolic 78.4 ± 0.69 78.1 ± 0.54 0.72
Hypertension prevalence* (%) 57 50 0.14

Data are means ± SD unless otherwise indicated.

*

Variables not adjusted for age, sex, ethnicity, and clinic;

by definition;

log-transformed and back-transformed for presentation. AA, African American; H, Hispanic; W, non-Hispanic white.

Table 3—

Clinical characteristics of insulin-resistant diabetic subjects by Si or Si > 0 (0 < Si < 1.61) adjusted for age, sex, ethnicity, clinic, and waist circumference

Si = 00 < Si < 1.61P
Cholesterol (mg/dl)
Total 210.7 ± 3.4 216.1 ± 2.8 0.210
HDL 38.9 ± 0.8 40.5 ± 0.6 0.120
LDL 136.4 ± 2.8 158.7 ± 1.0 0.05
Triglyceride (mg/dl) 212.9 ± 11.8 180.1 ± 9.2 0.19
LDL size (Å) 256.4 ± 0.7 257.5 ± 0.6 0.22
PAI-1 (ng/ml) 36.9 ± 1.7 30.3 ± 1.3 0.004
Fibrinogen (mg/dl) 299.9 ± 4.5 290.0 ± 3.5 0.15
CRP 3.56 ± 0.26 2.94 ± 0.17 0.035
Glucose (mg/dl)
Fasting 171.7 ± 4.6 175.4 ± 3.6 0.53
2-h 314.9 ± 6.8 312.7 ± 5.3 0.80
Insulin (μU/ml)
Fasting 28.6 ± 1.1 21.6 ± 0.9 <0.001
2-h 87.6 ± 0.1 64.1 ± 1.1 <0.001
Acute insulin response (μU · ml−1 · min−17.1 ± 1.5 7.0 ± 1.2 0.25
Blood pressure (mmHg)
Systolic 127.2 ± 0.91 126.9 ± 1.2 0.19
Diastolic 77.9 ± 0.69 78.2 ± 0.54 0.28
Hypertension prevalence (%) 67.2 58.7 0.09
Si = 00 < Si < 1.61P
Cholesterol (mg/dl)
Total 210.7 ± 3.4 216.1 ± 2.8 0.210
HDL 38.9 ± 0.8 40.5 ± 0.6 0.120
LDL 136.4 ± 2.8 158.7 ± 1.0 0.05
Triglyceride (mg/dl) 212.9 ± 11.8 180.1 ± 9.2 0.19
LDL size (Å) 256.4 ± 0.7 257.5 ± 0.6 0.22
PAI-1 (ng/ml) 36.9 ± 1.7 30.3 ± 1.3 0.004
Fibrinogen (mg/dl) 299.9 ± 4.5 290.0 ± 3.5 0.15
CRP 3.56 ± 0.26 2.94 ± 0.17 0.035
Glucose (mg/dl)
Fasting 171.7 ± 4.6 175.4 ± 3.6 0.53
2-h 314.9 ± 6.8 312.7 ± 5.3 0.80
Insulin (μU/ml)
Fasting 28.6 ± 1.1 21.6 ± 0.9 <0.001
2-h 87.6 ± 0.1 64.1 ± 1.1 <0.001
Acute insulin response (μU · ml−1 · min−17.1 ± 1.5 7.0 ± 1.2 0.25
Blood pressure (mmHg)
Systolic 127.2 ± 0.91 126.9 ± 1.2 0.19
Diastolic 77.9 ± 0.69 78.2 ± 0.54 0.28
Hypertension prevalence (%) 67.2 58.7 0.09

Data are means ± SD unless otherwise indicated.

Table 4—

Clinical characteristics of insulin-resistant subjects with IGT according to whether Si = 0 or Si > 0 (0 < Si < 1.61) adjusted for age, sex, ethnicity, and clinic

Si = 00 < Si < 1.61P
n 44 202
Age (years)* 51.5 ± 1.1 56.1 ± 0.8 0.724
Sex* (% female) 59 59 0.98
Ethnicity* (% W/AA/H) 39/27/34 34/29/38 0.82
BMI (kg/m234.3 ± 1.1 31.1 ± 0.5 0.007
Waist circumference (cm) 103.5 ± 2.0 97.0 ± 0.9 0.004
Waist-to-hip ratio 0.90 ± 0.01 0.88 ± 0.001 0.12
Cholesterol (mg/dl)
Total 207.7 ± 4.6 215.0 ± 2.6 0.17
HDL 38.8 ± 1.7 44.6 ± 1.0 0.004
LDL 144.2 ± 4.7 142.1 ± 2.6 0.70
Triglyceride (mg/dl) 160.1 ± 1.5 165.8 ± 7.0 0.71
LDL size (Å) 258.9 ± 0.7 259.7 ± 1.5 0.98
PAI-1 (ng/ml) 29.5 ± 2.6 28.2 ± 1.7 0.69
Fibrinogen (mg/dl) 314.2 ± 8.8 283.1 ± 4.0 0.002
CRP (mg/l) 2.94 ± 0.17 2.45 ± 0.20 0.001
Glucose (mg/dl)
Fasting 107.3 ± 1.6 105.3 ± 0.7 0.27
2-h 170.5 ± 2.7 164.6 ± 1.2 0.05
Insulin (μU/ml)
Fasting 26.5 ± 2.5 21.3 ± 1.6 0.08
2-h 250.0 ± 2.7 146.4 ± 7.0 <0.001
Acute insulin response (μU · ml−1 · min−150.7 ± 6.7 43.9 ± 3.2 0.36
Blood pressure (mmHg)
Systolic 125.7 ± 2.7 125.1 ± 1.3 0.85
Diastolic 80.4 ± 1.3 78.8 ± 0.63 0.28
Hypertension prevalence* (%) 41 47 0.50
Si = 00 < Si < 1.61P
n 44 202
Age (years)* 51.5 ± 1.1 56.1 ± 0.8 0.724
Sex* (% female) 59 59 0.98
Ethnicity* (% W/AA/H) 39/27/34 34/29/38 0.82
BMI (kg/m234.3 ± 1.1 31.1 ± 0.5 0.007
Waist circumference (cm) 103.5 ± 2.0 97.0 ± 0.9 0.004
Waist-to-hip ratio 0.90 ± 0.01 0.88 ± 0.001 0.12
Cholesterol (mg/dl)
Total 207.7 ± 4.6 215.0 ± 2.6 0.17
HDL 38.8 ± 1.7 44.6 ± 1.0 0.004
LDL 144.2 ± 4.7 142.1 ± 2.6 0.70
Triglyceride (mg/dl) 160.1 ± 1.5 165.8 ± 7.0 0.71
LDL size (Å) 258.9 ± 0.7 259.7 ± 1.5 0.98
PAI-1 (ng/ml) 29.5 ± 2.6 28.2 ± 1.7 0.69
Fibrinogen (mg/dl) 314.2 ± 8.8 283.1 ± 4.0 0.002
CRP (mg/l) 2.94 ± 0.17 2.45 ± 0.20 0.001
Glucose (mg/dl)
Fasting 107.3 ± 1.6 105.3 ± 0.7 0.27
2-h 170.5 ± 2.7 164.6 ± 1.2 0.05
Insulin (μU/ml)
Fasting 26.5 ± 2.5 21.3 ± 1.6 0.08
2-h 250.0 ± 2.7 146.4 ± 7.0 <0.001
Acute insulin response (μU · ml−1 · min−150.7 ± 6.7 43.9 ± 3.2 0.36
Blood pressure (mmHg)
Systolic 125.7 ± 2.7 125.1 ± 1.3 0.85
Diastolic 80.4 ± 1.3 78.8 ± 0.63 0.28
Hypertension prevalence* (%) 41 47 0.50

Data are means ± SD unless otherwise indicated.

*

Variables not adjusted for age, sex, ethnicity, and clinic. A, African American; H, Hispanic; W, non-Hispanic white.

Table 5—

Clinical characteristics of insulin-resistant subjects with IGT according to Si = 0 or Si > 0 (0 < Si < 1.61) adjusted for age, clinic, ethnicity, sex, and waist circumference

Si = 00 < Si < 1.61P
Age (years) 56.1 ± 0.6 57.5 ± 1.1 0.267
Cholesterol (mg/dl)
Total 207.7 ± 4.4 215.0 ± 2.6 0.171
HDL 38.8 ± 1.7 44.6 ± 1.0 0.004
LDL 144.2 ± 4.7 142.1 ± 36.2 0.662
Triglyceride (mg/dl) 165.8 ± 7.0 160.1 ± 13.6 0.707
LDL size (Å) 258.9 ± 0.7 259.7 ± 1.5 0.635
PAI-1 (ng/ml) 29.5 ± 2.6 28.2 ± 1.7 0.686
Fibrinogen (mg/dl) 314.2 ± 8.8 283.1 ± 4.0 0.002
CRP (mg/l) 3.70 ± 0.62 2.48 ± 0.19 0.019
Glucose (mg/dl)
Fasting 107.3 ± 1.6 109.5 ± 0.7 0.277
2-h 170.5 ± 2.7 164.6 ± 1.2 0.053
Insulin (μU/ml)
Fasting 26.5 ± 2.5 21.3 ± 1.6 0.081
2-h 250.0 ± 29.0 146.4 ± 7.0 0.001
Acute insulin response (μU · ml−1 · min−150.7 ± 6.7 43.9 ± 3.2 0.367
Blood pressure (mmHg)
Systolic 124.8 ± 2.7 125.3 ± 1.28 0.86
Diastolic 79.8 ± 1.3 78.8 ± 0.63 0.51
Hypertension prevalence (%) 62.2 49.7 0.18
Si = 00 < Si < 1.61P
Age (years) 56.1 ± 0.6 57.5 ± 1.1 0.267
Cholesterol (mg/dl)
Total 207.7 ± 4.4 215.0 ± 2.6 0.171
HDL 38.8 ± 1.7 44.6 ± 1.0 0.004
LDL 144.2 ± 4.7 142.1 ± 36.2 0.662
Triglyceride (mg/dl) 165.8 ± 7.0 160.1 ± 13.6 0.707
LDL size (Å) 258.9 ± 0.7 259.7 ± 1.5 0.635
PAI-1 (ng/ml) 29.5 ± 2.6 28.2 ± 1.7 0.686
Fibrinogen (mg/dl) 314.2 ± 8.8 283.1 ± 4.0 0.002
CRP (mg/l) 3.70 ± 0.62 2.48 ± 0.19 0.019
Glucose (mg/dl)
Fasting 107.3 ± 1.6 109.5 ± 0.7 0.277
2-h 170.5 ± 2.7 164.6 ± 1.2 0.053
Insulin (μU/ml)
Fasting 26.5 ± 2.5 21.3 ± 1.6 0.081
2-h 250.0 ± 29.0 146.4 ± 7.0 0.001
Acute insulin response (μU · ml−1 · min−150.7 ± 6.7 43.9 ± 3.2 0.367
Blood pressure (mmHg)
Systolic 124.8 ± 2.7 125.3 ± 1.28 0.86
Diastolic 79.8 ± 1.3 78.8 ± 0.63 0.51
Hypertension prevalence (%) 62.2 49.7 0.18

Data are means ± SD unless otherwise indicated.

This work was supported by National Heart, Lung, and Blood Institute Grants HL47887, HL47889, HL47890, HL47892, and HL47902 and the General Clinical Research Centers Program (NCRR GCRC, M01 RR431, and M01 RR01346).

1.
Sicree RA, Zimmet PZ, King HOM, Coventry JS: Plasma insulin response among Nauruans: Prediction of deterioration in glucose tolerance over 6 years.
Diabetes
36
:
179
–186,
1987
2.
Haffner SM, Stern MP, Mitchell BD, Hazuda HP, Patterson JK: Incidence of type II diabetes in Mexican Americans predicted by fasting insulin and glucose levels, obesity, and body fat distribution.
Diabetes
39
:
283
–288,
1990
3.
Bergstrom RW, Newell-Morris LL, Leonetti DL, Shuman WP, Wahl PW, Fujimoto WY: Association of elevated fasting C-peptide level and increased intra-abdominal fat distribution with development of NIDDM in Japanese-American men.
Diabetes
39
:
104
–111,
1990
4.
Charles MA, Fontbonne A, Thibult N, Warnet JM, Rosselin GE, Eschwege E: Risk factors for NIDDM in white populations: Paris Prospective Study.
Diabetes
49
:
796
–799,
1991
5.
Mykkänen L, Kuusisto J, Pyörälä K, Laakso M: Cardiovascular disease risk factors as predictors of type II (non-insulin-dependent) diabetes mellitus in elderly subjects.
Diabetologia
36
:
553
–559,
1993
6.
Warram JH, Martin BC, Krolewski AS, Soeldner JS, Kahn CR: Slow glucose removal rate and hyperinsulinemia precede the development of type II diabetes in the offspring of diabetic parents.
Ann Intern Med
113
:
909
–915,
1990
7.
Lillioja S, Mott DM, Spraul M, Ferraro R, Foley JE, Ravussin E, Knowler WC, Bennett PH, Bogardus C: Insulin resistance and insulin secretory dysfunction as precursors of non-insulin dependent diabetes mellitus: prospective studies of Pima Indians.
N Engl J Med
329
:
1988
–1992,
1993
8.
Reaven GM: Banting Lecture: role of insulin resistance in human disease.
Diabetes
37
:
1595
–1607,
1988
9.
Haffner SM, Valdez RA, Hazuda HP, Mitchell BD, Morales PA, Stern MP: Prospective analyses of the Insulin Resistance Syndrome (Syndrome X).
Diabetes
41
:
715
–722,
1992
10.
Zavaroni I, Bonora E, Pagliara M, Dall’Aglio E, Luchetti L, Buonanno G, Bonati PA, Berganzoni M, Gnudi L, Passeri M, Reaven GM: Risk factors for coronary artery disease in healthy persons with hyperinsulinemia and normal glucose tolerance.
N Engl J Med
320
:
702
–706,
1989
11.
Laakso M, Sarlund H, Mykkänen L: Insulin resistance is associated with lipid and lipoprotein abnormalities in subjects with varying degrees of glucose tolerance.
Arteriosclerosis
10
:
223
–231,
1989
12.
Ferrannini E, Buzzigoli G, Bonadonna R, Giorico MA, Oleggini M, Graziedi L, Pedrinelli R, Brandi L, Bevilacqua S: Insulin resistance in essential hypertension.
N Engl J Med
317
:
350
–357,
1987
13.
Vague P, Juhan-Vague I, Aillaud MF, Badier C, Viard R, Alessi MC, Collen D: Correlation between fibrinolytic activity, plasminogen activator inhibitor level, plasma insulin level and relative body weight in normal and obese subjects.
Metabolism
35
:
250
–253,
1986
14.
Mykkänen L, Rönnemaa T, Marniemi J, Haffner SM, Bergman R, Laakso M: Insulin sensitivity is not an independent determinant of plasma plasminogen activator inhibitor-1.
Arterioscler Thromb
14
:
1264
–1271,
1994
15.
Pyörälä K, Savolainen E, Kaukola S, Haapakoski J: Plasma insulin as coronary heart disease risk factor: relationship to other risk factors and predictive during 9
$${1}/{2}$$
year follow-up of the Helsinki Policeman Study population.
Acta Med Scand
701 (Suppl.)
:
38
–52,
1985
16.
Després JP, Lamarche B, Mauriége P, Cantin B, Dagenais GR, Moorjani S, Lupien PJ: Hyperinsulinemia as an independent risk factor for ischemic heart disease.
N Engl J Med
334
:
952
–957,
1996
17.
Howard G, O’Leary DH, Zaccaro D, Haffner SM, Rewers M, Hamman R, Selby JV, Saad MF, Savage P, Bergman R: Insulin sensitivity and atherosclerosis.
Circulation
93
:
1809
–1817,
1996
18.
Laakso M: How good a marker is insulin level for insulin resistance?
Am J Epidemiol
137
:
959
–965,
1993
19.
DeFronzo RA, Tobin JD, Andres R: Glucose clamp technique: a method for quantifying insulin secretion and resistance.
Am J Physiol
237
:
E214
–E223,
1979
20.
Bergman RN, Ider YZ, Bowden CR, Cobelli C: Quantitative estimation of insulin sensitivity.
Am J Physiol
236
:
E667
–E677,
1979
21.
Yang Y, Yon JH, Bergman RN: Modified protocols improve insulin sensitivity estimation using the minimal model.
Am J Physiol
253
:
E595
–E603,
2987
22.
Welch S, Gebhart SSP, Bergman RN, Phillips LS: Minimal model analysis of intravenous glucose tolerance test derived insulin sensitivity in diabetic subjects.
J Clin Endocrinol Metab
71
:
1508
–1518,
1990
23.
Steil GM, Volund A, Kahn SE, Bergman RN: Reduced sample number for calculation of insulin sensitivity and glucose effectiveness from the minimal model: suitability for use in population studies.
Diabetes
42
:
250
–256,
1993
24.
Saad MF, Anderson RL, Laws A, Watanabe RM, Kades WW, Chen YDI, Sands RE, Pei D, Savage PJ, Bergman RN: A comparison between the minimal model and the glucose clamp in the assessment of insulin sensitivity across the spectrum of glucose tolerance.
Diabetes
43
:
1114
–1121,
1994
25.
Cobelli C, Pacini G, Toffolo G, Sacca L: Estimation of insulin sensitivity and glucose clearance from minimal model: new insights from labeled IVGTT.
Am J Physiol
250
:
E591
–E598,
1986
26.
Ni T-C, Ader M, Bergman RN: Reassessment of glucose effectiveness and insulin sensitivity from minimal model analysis: a theoretical evaluation of the single compartment glucose distribution assumption.
Diabetes
46
:
1813
–1821,
1997
27.
Finegood DT, Hramiak IM, Dupre J: A modified protocol for estimation of insulin sensitivity with the minimal model of glucose kinetics in patients with insulin-dependent diabetes.
J Clin Endocrinol Metab
70
:
1438
–1549,
1990
28.
Doberne L, Greenfield MS, Schulz B, Reaven GM: Enhanced glucose utilization during prolonged glucose clamp studies.
Diabetes
30
:
829
–835,
1981
29.
Haffner SM, D’Agostino R Jr, Mykkänen L, Tracy R, Howard B, Rewers M, Selby J, Savage PJ, Saad MF: Insulin sensitivity in subjects with type 2 diabetes: relationship to cardiovascular risk factors: the Insulin Resistance Atherosclerosis Study.
Diabetes Care
22
:
562
–568,
1999
30.
Wagenknecht LE, Mayer EJ, Rewers M, Haffner SM, Selby J, Burke GM, Henkin L, Howard G, Savage PJ, Saad MF, Bergman RN, Hamman R: The Insulin Resistance Atherosclerosis Study (IRAS): objectives, design and recruitment results.
Ann Epidemiol
5
:
464
–471,
1995
31.
Haffner SM, D’Agostino R Jr, Saad MF, Rewers M, Mykkänen L, Selby J, Howard G, Savage PJ, Hamman RF, Wagenknecht LE, Bergman RN: Increased insulin resistance and insulin secretion in nondiabetic African-Americans and Hispanics compared with non-Hispanic whites: the Insulin Resistance Atherosclerosis Study.
Diabetes
45
:
742
–748,
1996
32.
Haffner SM, Howard G, Mayer E, Bergman RN, Savage P, Rewers M, Mykkänen L, Selby JV, Saad MF: Insulin sensitivity and acute insulin response in African American, non-Hispanic whites and Hispanic subjects with NIDDM: the Insulin Resistance Atherosclerosis Study.
Diabetes
46
:
63
–69,
1997
33.
World Health Organization:
Diabetes Mellitus: Report of a WHO Study Group
. Geneva, World Health Org.,
1985
(Tech. Rep. Ser., no. 727)
34.
Pacini G, Bergman RN: MINMOD: a computer program to calculate insulin sensitivity and pancreatic responsibility from the frequently sampling intravenous glucose tolerance test.
Comput Methods Programs Biomed
23
:
113
–122,
1986
35.
Krauss RM, Burke DJ: Identification of multiple subclasses of plasma low density lipoproteins in normal humans.
J Lipid Res
23
:
97
–104,
1982
36.
Haffner SM, Mykkänen L, Stern MP, Paidi M, Howard BV: Greater effect of diabetes on LDL size in women than in men.
Diabetes Care
17
:
1164
–1171,
1994
37.
Geffken D, Keating F, Kennedy M, Cornell E, Bovill E, Tracy R: The measurement of fibrinogen in population-based research: studies on instrumentation and methodology.
Arch Pathol Lab Med
118
:
1106
–1109,
1994
38.
Clauss A: Gerinnungs-physiologische schnell-methode zur bestimmung des fibrinogens.
Acta Haematol
17
:
237
–246,
1957
39.
Macy E, Meilahn E, Declerck P, Tracy R: Sample preparation for plasma measurement of plasminogen activator inhibitor-1 antigen in large population studies.
Arch Path Lab Med
117
:
67
–70,
1993
40.
Declerck P, Collen D: Measurement of plasminogen activator inhibitor 1 (PAI-1) in plasma with various monoclonal antibody-based enzyme-linked immunosorbent assays.
Thromb Res Suppl
10
:
3
–9,
1990
41.
Macy EM, Hayes TE, Tracy RP: Variability in the measurement of C-reactive protein in healthy subjects: implications for reference intervals and epidemiological applications.
Clin Chem
43
:
52
–58,
1997
42.
Pouliot MC, Despré JP, Lemieux S, Moorjani S, Bouchard C, Tremblay A, Nadeau A, Lupien PJ: Waist circumference and abdominal sagittal diabetes: best simple anthropometric indexes of abdominal visceral adipose tissue accumulation and related cardiovascular risk in men and women.
Am J Cardiol
73
:
460
–468,
1994
43.
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
285
:
2486
–2497,
2001
44.
Alzaid AA, Dinneen SF, Turk DJ, Caumo A, Cobelli C, Rizza RA: Assessment of insulin action and glucose effectiveness in diabetic and nondiabetic humans.
J Clin Invest
6
:
2341
–2348,
1994
45.
Expert Panel on Detection, Evaluation, and Treatment of High Cholesterol in Adults (Adult Treatment Panel III): Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP).
JAMA
285
:
2486
–2497,
2001

A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances.