Impact of Traits of Metabolic Syndrome on β-Cell Function and Insulin Resistance in Normal Fasting,Normal Glucose Tolerant Subjects

Abstract

Objective:

Metabolic syndrome, impaired fasting glucose (IFG), and impaired glucose tolerance (IGT) predict risk for type 2 diabetes mellitus (T2DM). To determine if increased risk preceded development of these abnormalities, β-cell function and insulin resistance were assessed in euglycemic subjects with and without traits of metabolic syndrome.

Methods:

A total of 562 apparently healthy Latin-American subjects were screened for metabolic syndrome [National Education Cholesterol Program Adult Treatment Panel III (NECP ATP III)]. Early pancreatic insulin response ΔInsulin_0–30/ΔGlucose_0–30, Matsuda index, disposition index (DI), and homeostasis model assessment of insulin resistance (HOMA-IR) ratio were obtained from oral glucose tolerance testing (0–180 min).

Results:

ΔI_0–30/ΔG_0–30, Matsuda index, DI, and HOMA-IR deteriorated in direct proportion with number of traits of metabolic syndrome, and with increases in glucose levels within the euglycemic range. DI was the most sensitive index. In subjects with 1, 2, 3, and 4–5 traits, DI was 21.4%, 40%, 57%, and 76% lower, respectively, than in subjects with no traits. As a single trait, abdominal obesity was associated with insulin resistance, whereas, low high-density lipoprotein cholesterol (HDL-C), alone or combined with high triglycerides, was not associated with insulin resistance or β-cell dysfunction. Combined impairments in β-cell function and insulin sensitivity were responsible for the increases in fasting and 2-h plasma glucose concentrations within the euglycemic range.

Conclusions:

Impaired β-cell function and increased insulin resistance are present much before development of metabolic syndrome, IFG, or IGT. β-Cell function and insulin sensitivity worsen in direct proportion with number of traits of metabolic syndrome and increases in glucose levels. Compared to abdominal obesity, low HDL-C±high triglycerides may bear a lesser weight in predicting risk of T2DM.

Introduction

I dentification of subjects at high risk of developing type 2 diabetes mellitus (T2DM) is required for early initiation of preventive treatment. Impaired fasting glucose (IFG) and impaired glucose tolerance (IGT) are glucose abnormalities known to be strongly associated with a higher future risk of T2DM.^1,2 However, a significant number of euglycemic individuals are also at increased risk of developing T2DM, which limits the use of IFG and IGT as the sole means of identifying subjects at high risk.^1

–4 Unfortunately, these subjects are not identified when adhering to current guidelines. In addition to glucose abnormalities, the presence of metabolic syndrome also predicts increased risk of T2DM.^5

–8 Again, some subjects with one or two traits not meeting criteria for metabolic syndrome may also be at increased risk of developing T2DM.³ On the basis of these observations, we propose that early abnormalities in β-cell function and insulin resistance not detected by routine screening ought to account for the increased risk of future T2DM, observed in some euglycemic subjects.

Defects in β-cell function and in insulin sensitivity play a pathogenic role in T2DM.^9,10 Identification of abnormalities in insulin secretion and sensitivity, preceding significant changes in fasting and postload glucose levels, may help in the early detection of subjects at high risk for T2DM. Previous studies have shown that the early insulin response to a glucose load (ΔI_0–30/ΔG_0–30), also known as early pancreatic response or insulinogenic index, is a reliable surrogate of β-cell function. This index may be sensitive enough to detect early impairments in β-cell function because it has been shown to identify normal fasting–normal glucose tolerant (NFG-NGT) subjects that are at high risk for T2DM.^11

–14 Whole-body insulin sensitivity has been estimated with the Matsuda index, which is calculated from the relative changes in insulin and glucose levels at 0, 30, 60, 120, and 180 min during an oral glucose tolerance test (OGTT).^13,14 Of all indices, the disposition index (insulin secretion/insulin resistance index), which takes into account both β-cell function and insulin sensitivity, has been shown to be a better predictor of risk of developing T2DM than either of its components.^15
–17 Additionally, homeostasis model assessment of insulin resistance (HOMA-IR), which is based on fasting glucose and insulin values, and reflects mainly hepatic insulin resistance, has also been used to assess insulin resistance.^11
–13

In this study, the above-described surrogate markers of β-cell function and insulin sensitivity were employed to screen for early abnormalities in insulin-glucose metabolism, which may be linked to increased risk for T2DM. We determined if increases in fasting and postload plasma glucose concentrations within the euglycemic range were characterized and perhaps determined by impairments in β-cell function and/or insulin sensitivity. In addition, we investigated whether the presence of traits of metabolic syndrome, as well as concurrence of its components, may also mark early defects in insulin resistance and β-cell function. Values for early insulin response, Matsuda and disposition indices, and HOMA-IR ratio were obtained and compared in subjects with no traits, one, two, three, four, or five traits of the metabolic syndrome. National Education Cholesterol Program Adult Treatment Panel III (NECP ATP III) guidelines for diagnosis of metabolic syndrome were employed. Noteworthy, these guidelines give equal weight to each of the traits or components.¹⁸ This view is challenged by the International Diabetes Federation (IDF) guidelines, where abdominal obesity is a necessary trait for diagnosis of metabolic syndrome.^19,20 In this study, we determined whether low high-density lipoprotein cholesterol (HDL-C), high triglycerides, high blood pressure, and abdominal obesity when present as a single trait, showed similar degrees of impairment of insulin resistance and β-cell function, in NFG-NGT subjects.

Methods

This study was performed at the Center for the Detection of Silent Cardiovascular and Metabolic risk factors affiliated with the Clinical Pharmacology Unit at the Central University of Venezuela. A total of 562 apparently healthy subjects were screened for glucose abnormalities and cardiovascular and metabolic risk factors. Patients with known type 1 DM were excluded. None of the studied subjects was taking medications affecting glucose or insulin metabolism. The study was conducted in adherence to the Declaration of Helsinki, and the research protocol was approved by the institutional review board of the Central University Hospital serving the city of Caracas. All participants gave written informed consent. All applicable institutional and governmental regulations concerning the ethical use of human volunteers were followed during this research.

Complete history, physical examination, and laboratory investigations, including hematology, chemistry, fasting lipid panel, fasting and postload (75 g d-glucose) glucose and insulin levels, liver function tests, and urinalysis, were obtained. Blood pressure was measured with a standard mercury sphygmomanometer, and the cuff size was optimized for arm circumference. Heart rate was obtained from a 1-min pulse. Overall adiposity was assessed by body weight and body mass index (BMI). Waist circumference was measured in the standing position midway between the highest point of the iliac crest and the lowest point of the costal margin in the midaxillary line. All anthropometric measurements reflected the average of two measurements.

After at least 5 days of weight-maintaining diet, the fasting subjects underwent a 75-g OGTT. Blood samples were obtained at baseline, 30, 60, 90, 120, and 180 min after the glucose ingestion. Patients were classified into groups based on the American Diabetes Association (ADA) and World Health Organization (WHO)^2
–4 diagnostic criteria based on glucose levels: Normal glucose tolerance (fasting <100 mg/dL and 2-h <140 mg/dL); isolated IFG (fasting 100–125 mg/dL and 2 h <140 mg/dL); isolated IGT (fasting <100 mg/dL and 2 h 140–199 mg/dL); and combined IFG/IGT (fasting 100–125 mg/dL and 2 h 140–199 mg/dL. T2DM was diagnosed with fasting glucose >125 mg/dL and/or 2-h postload glucose >199 mg/dL. All subjects were then classified based on 1-h postload plasma glucose concentrations as those with less than 155 mg/dL and ≥155 mg/dL.

HOMA-IR, an index of hepatic insulin resistance, was calculated as (fasting insulin μUI/mL×glucose mmol/L)/22.5.⁹ The total insulin response (ΔInsulin_0–180 /ΔGlucose_0–180 ratio) and the early insulin response (ΔInsulin_0–30/ΔGlucose_0–30 ratio), also known as insulinogenic indices, were calculated.^10,11 The Matsuda index, an approximation of whole-body insulin sensitivity¹³ and the disposition index (ΔI_0–30/ΔG_0–30 ÷ Matsuda index) strong predictors of risk of diabetes,^13,14 were also determined.

Subjects were categorized as having the metabolic syndrome if they met at least three of the NCEP ATP III criteria: Waist circumference >102 cm (>40 in) in men and >88 cm (35 in) in women; triglycerides ≥150 mg/dL (≥1.7 mmol/L), HDL-C <40 mg/dL (<1.03 mmol/lt) in men and <50 mg/dL (≥1.29 mmol/L) in women, blood pressure ≥130/≥85 mmHg or on antihypertensive medication, and fasting glucose ≥110 mg/dL (≥6.1 mmol/lt).¹⁶

Blood samples for plasma glucose and insulin were obtained from an arm vein. Five-milliliter samples were collected in heparinized ice-cold tubes, mixed by gentle inversion, and immediately centrifuged at 4°C for 15 min. The plasma was divided in three aliquots, two of which were stored at −40°C for subsequent insulin assay and the other immediately employed for glucose quantification. Plasma glucose was measured with an automated glucose analyzer (Beckman Instruments, Palo Alto, CA), employing a glucose oxidase technique. The assay was linear between 5 and 500 mg/dL of plasma glucose. The intraassay variation was of 1.7% and the interassay variation was of 2.2%. Plasma insulin was quantitated by radioimmunoassay solid-phase radioimmunoassay (Immulite^® 2000 Insulin, Diagnostic Products Corporation, Los Angeles, CA). Intraassay and interassay variability were 5.2% and 8.1%, respectively. The assay sensitivity was 2 μUI/mL, and the antibody was 100% specific for insulin. The assay showed linearity up to 290 μUI/mL of plasma insulin.

Statistical analyses

Descriptive statistics were generated for the study population using means [and standard error of the mean (SEM)] for continuous variables and proportions (%) for dichotomous variables. The significance of mean differences was tested with analysis of variance (ANOVA). Differences between categorical variables were tested with the chi-squared test. Two-sample comparison for continuous variables was analyzed with the Student t-test or paired t-test with Bonferroni adjustment for repeated testing. Triglyceride levels were log-transformed for statistical analysis and back-transformed for reporting. Pearson correlation analysis was used to assess the relationships between continuous variables. Differences were considered significant at values of P<0.05. All statistical analysis was performed with SPSS version 11.0 (SPSS Inc, Chicago, IL).

Results

A total of 562 unselected subjects were evaluated for insulin and glucose abnormalities and traits of the metabolic syndrome. Plasma glucose and insulin concentrations after a 75-g glucose oral load were measured as part of the 0- to 180-min OGTT. β-cell function estimated by the early pancreatic response (insulinogenic index or ΔI_0–30/ΔG_0–30), the Matsuda index (an estimator of whole body insulin resistance), the disposition index (a combination of both previous indices), and the HOMA-IR index (an estimator of hepatic insulin resistance) were calculated as described under Methods.

Early abnormalities in β-cell function and insulin sensitivity in euglycemic subjects

To determine if impairments of insulin secretion and/or sensitivity are already present in subjects with fasting glucose concentrations in the normal range, the four indices were measured in subjects with fasting serum glucose levels lower than 90 mg/dL, between 90 and 99 mg/dL, 100–125 mg/dL, and 110–125 mg/dL. All four indices showed gradual deterioration with increasing fasting plasma glucose (FPG) concentrations (Table 1). Compared to subjects with FPG concentrations lower than 90 mg/dL, those with levels in the upper euglycemic range (90–99 mg/dL) showed reductions (∼20%) in early pancreatic response and in the Matsuda index, a 37% reduction in the disposition index, and a 28% increase in HOMA-IR. Expectedly, IFG (FPG concentrations between 100 and 125 mg/dL) was associated with further deterioration of all four indices. Further impairment was observed if the cutoff for FPG concentrations was increased from 100 mg/dL to 110 mg/dL (Table 1). Of all indices, the disposition index showed the greater percentage change, suggesting that increases in FPG concentrations result from combined impairments in β-cell function and insulin sensitivity.

Table 1.

Changes in Insulinogenic Index, Homestasis Model Assessment of Insulin Resistance, Matsuda Index, and Disposition Index in Subjects with Normal Fasting Glucose and Impaired Fasting Glucose

Fasting serum glucose (mg/dL)	Insulinogenic index (ΔI_0–30 /ΔG_0–30)	HOMA-IR	Matsuda index	Disposition index
<90 (n=260)	1.8±0.15	2.55±0.12	3.5±0.15	6.3±0.4
90–99 (n=166)	1.43±0.11^*	3.15±0.11^*	2.8±0.1^*	4.0±0.2^*
>100 to <126 (n=136)	1.11±0.12^*	4.22±0.23^*	2.02±0.14^*	2.25±0.2^*
>110 (n=42)	0.8±0.1^*	4.83±0.41^*	1.81±0.11^*	1.44±0.18^*

Shown are mean values±standard error of the mean (SEM).

Significantly different from subjects with fasting plasma glucose concentrations below 90 mg/dL at P<0.01.

HOMA-IR, homeostasis model assessment of insulin resistance.

To investigate if significant impairments in insulin secretion and/or sensitivity were present prior to the development of IGT, the four indices were measured in subjects with 2-h postload glucose concentrations lower than 120 mg/dL, within 120 and 139 mg/dL, and in subjects considered as IGT (140–199 mg/dL) (Table 2). When compared to subjects with 2-h postload glucose levels of <120 mg/dL, those with 2-h plasma glucose concentrations between 120 and 139 mg/dL showed significant alterations in the four indices, suggesting the presence of abnormalities in insulin resistance and in β-cell function, despite being in the euglycemic range. The disposition index was 45% lower in the <120–139 mg/dL than in the <120 mg/dL group. As expected, further deterioration of the four indices was observed in IGT. Compared to NGT subjects, subjects with IGT had a 47% reduction of the insulinogenic index, 35% reduction of the Matsuda index, and 65% reduction of the disposition index. The HOMA-IR index was 40% higher in IGT than in NGT (Table 2).

Table 2.

Changes in Insulinogenic Index, Homeostasis Model Assessment of Insulin Resistance, Matsuda Index, and Disposition Index in Subjects with Normal Fasting Glucose and Impaired Fasting Glucose

2-h postload glucose (mg/dL)	Insulinogenic index	HOMA-IR	Matsuda index	Disposition index
<120 (n=331)	1.72±0.12	2.35±0.2	4.01±0.23	6.91±0.4
120–139 (n=82)	1.51±0.11	3.14±0.18^*	3.53±0.21^*	3.81±0.3^*
>140–199 (n=86)	0.87±0.2^*	3.74±0.22^*	2.34±0.18^*	2.03±0.21^*

Shown are mean values±standard error of the mean (SEM).

Significantly different from subjects with 2-h postload plasma glucose concentrations lower than 120 mg/dL at P<0.01.

HOMA-IR, homeostasis model assessment of insulin resistance.

In summary, gradual increases in fasting and 2-h postload plasma glucose concentrations within the fasting and postload euglycemic range seem to result from gradual deterioration of both β-cell function and insulin sensitivity.

Impact of traits of metabolic syndrome on β-cell function and insulin sensitivity

The early pancreatic insulin response (ΔI_0–30/ΔG_0–30), a surrogate marker of β-cell function, was measured in subjects with no traits of the metabolic syndrome, as well as in subjects with one, two, three, four, or five traits. Traits of metabolic syndrome were defined based on NECP ATP III guidelines. The early pancreatic insulin response was indirectly related to the number of traits of the metabolic syndrome concurring in a subject (Fig. 1). Subjects with advanced syndrome had 60% lower indices than subjects with no traits (P<0.01), and also 40% lower than subjects with three traits (P<0.01). These results suggest that impairment of β-cell function, assessed by the early pancreatic insulin response to a glucose load, is directly related to the number of traits of metabolic syndrome.

FIG. 1.

Early insulin abnormalities and traits of the metabolic syndrome. Shown are the relationship between number of traits or components of the metabolic syndrome present in an individual, as defined by National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III), and the values for indices of insulin sensitivity and resistance obtained from serial measurements of plasma insulin and glucose during an oral glucose tolerance test (OGGT). The following indices were obtained: Insulinogenic index, also known as the early pancreatic insulin response to an oral glucose load, was calculated as ΔI_0–30/ΔG_0–30; Matsuda index, an indicator of insulin sensitivity was calculated as described in (13); disposition index, calculated as: (ΔI_0–30/ΔG_0–30)×Matsuda index; and the homeostasis model assessment of insulin resistance (HOMA-IR) was calculated as: (fasting insulin μUI/mL×glucose mg/dL)/405. Shown are mean values±standard error of the mean (SEM). MS, Metabolic syndrome.

Surrogate markers of insulin resistance such as HOMA-IR and the Matsuda index, based either on fasting plasma glucose and insulin levels (HOMA-IR) or on relative glucose and insulin concentrations during the OGTT, also showed gradual impairments with increased number of traits (Fig. 1). HOMA-IR increased with number of traits of the metabolic syndrome. HOMA-IR was 10%, 23%, and 45% greater in subjects with one, two, or three traits, respectively, compared to subjects with no traits. HOMA-IR values were 150% higher in subjects with four to five traits than in those with no traits. For subjects with diagnosis of metabolic syndrome, HOMA-IR values were 50% greater in subjects with four to five traits than in subjects with three traits. Similarly, the Matsuda index was 10%, 20%, and 33% greater in subjects with one, two, or three traits, respectively, compared to subjects with no traits. The Matsuda index showed a 60% decrease in subjects with four to five traits compared to those with no traits (P<0.01); in addition, subjects with four to five traits had 32% lower values for the Matsuda index than those with three traits (P<0.01) (Fig. 1).

The combination of defects in insulin sensitivity (Matsuda index) and early insulin secretion (insulinogenic index) measured as the disposition index, was found to be a more sensitive index of impaired insulin–glucose metabolism than either of its components. The disposition index declined progressively with the increase in number of traits. Subjects with one, two, three, and four to five traits had values for the disposition index that were 21.4%, 40%, 57%, and 76% lower, respectively, than the values obtained in subjects with no traits (Fig. 1).

In summary, gradual impairments in β-cell function and insulin sensitivity were observed with an increase number of traits of metabolic syndrome; such impairments were already present in subjects with one and two traits not meeting criteria of metabolic syndrome. Furthermore, subjects with metabolic syndrome but with four or five traits had greater impairment in β-cell function and insulin sensitivity than those with three traits.

Type of trait of metabolic syndrome and impairments in β-cell function and insulin sensitivity

To determine which single trait was associated with the most impairment in insulin secretion and in insulin sensitivity, we quantified the four indices in euglycemic-normotensive subjects with one trait of metabolic syndrome. Low HDL-C and abdominal obesity were the most common single traits found in our study subjects (Table 2). There were only 4 subjects with high triglyceride levels and 6 subjects with high blood pressure as the sole trait; therefore, these groups were not included in the single-trait analysis. Data for nonobese, normotensive, euglycemic subjects with high triglycerides and low HDL-C (two traits) were also analyzed.

Body weight, waist circumference, waist-to-hip ratio, BMI, systolic blood pressure (SBP), diastolic blood pressure (DBP), fasting serum glucose and insulin, and 2-h insulin levels of subjects with low HDL-C as a single trait, and low-HDL-C+high triglycerides (two traits) were not significantly different from those of subjects with no traits (Table 3). However, subjects with abdominal obesity as a single trait had higher blood pressure and higher fasting and 2-h plasma insulin concentrations than subjects with no traits, low HDL-C as a single trait, or low HDL-C plus high triglycerides as combined traits. The sole presence of abdominal obesity was characterized by a 21% increase in HOMA-IR, and a 20% and 25% decrease in Matsuda and disposition indices, respectively, without significant changes in the early pancreatic response (insulinogenic index) (Table 3). Additional worsening of the Matsuda index, disposition index, and HOMA-IR values was observed if the group of subjects had three traits, namely, abdominal obesity, with low HDL-C and high triglycerides, and thus met the criteria for metabolic syndrome (Table 3).

Table 3.

Differential Effects of Abdominal Obesity and of Decreased High-Density Lipoprotein Cholesterol Levels on Surrogate Markers of Insulin Secretion and Insulin Sensitivity

	No traits	One trait: Obesity	One trait: Low HDL-C	Two traits: Low HDL-C+high TG	Three Traits: Obesity+low HDL-C+high TG
n	39	29	58	44	40
Age (years)	36.9±2	41.5±2^*	36.3±1.7	38.8±1.9	42.9±1.7^*
Weight (kg)	65.8±1.5	87±2^*	65.4±1.2	64.3±1.4	84±2.1^*
Waist	82.1±1.5	103±1.7^*	81.5±1	81.4±1.2	102.3±1.8^*
WHR	0.86±0.01	0.929±0.01^*	0.83±0.01	0.85±0.01	0.92±0.01^*
BMI	25.2±0.5	32±0.7^*	25.3±0.4	25.2±0.5	32.2±0.7^*
SBP	113±1.7	118±0.2^*	111±1	112±1.6	118±1.8^*
DBP	73.3±1.3	77±1.3^*	72±1	72±1.1	76±1.2^*
FSG	87±1.1	89±1.4	86±1	86±1.6	86±1.7
FSI	10.2±1.5	16±2^*	10.2±0.7	9.5±1	16.1±1.5^*
2-h G	93±2.7	106±3^*	99±3	104±4	118.7±8
2-h I	58±9	75±8^*	60±6	60±5	89±9
HDL	57.4±1.8	57±2	37±0.9^*	32±1^*	35.3±1.3^*
TG	81.5±5	91±5	95±4	242±14^*	218±13^*
(ΔI_0–30 /ΔG_0–30)	1.51±0.3	1.43±0.2	1.61±0.2	1.50±0.2	1.51±0.2
HOMA-IR	2.19±0.2	2.64±0.2^*	2.27±0.1	2.27±0.2	3.2±0.35^*
Matsuda index	4.11±0.3	3.29±0.3^*	4.30±0.3	3.88±0.4	2.79±0.4^*
Disposition index	6.19±0.5	4.72±0.33^*	6.93±0.6	5.82±0.5	3.16±0.5^*

Shown are mean values±standard error of the mean (SEM).

Significantly different from subjects with no traits at P<0.01.

WHR, waist-to-hip ratio; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; FSG, fasting serum glucose; FSI, fasting serum insulin; 2-h G, 2-hour glucose; 2-h I, 2-h insulin; HDL, high-density lipoprotein; TG, triglycerides; ΔI_0–30 /ΔG_0–30, early pancreatic response; HOMA-IR, homeostasis model assessment of insulin resistance.

In summary, these results suggest that abdominal obesity, even as a single trait, is associated with a state of insulin resistance without significant impairment in β-cell function. Low HDL-C as a single trait, or even combined with high triglycerides, was not found to be associated with significant impairment in β-cell function and/or in insulin sensitivity.

Discussion

Predicting future risk for T2DM is useful in initiating and designing early intervention strategies. Presence of fasting (IFG) and postload (IGT) glucose abnormalities, also known as prediabetes, mark increased risk for developing T2DM.^2
–4 However, not all subjects with IFG or IGT are at the same risk,^2
–4 and, in addition, some subjects diagnosed by NECP ATP III, ADA, and WHO guidelines as NFG-NGT, may also be at increased risk for future T2DM.^1,2 By employing surrogate markers of early pancreatic insulin response and of insulin sensitivity, it was possible to detect a progressive impairment of glucose homeostasis, much before fasting and postload hyperglycemia reached IFG and IGT levels (present study). Even within the normal range for fasting glucose concentrations, gradual increases in fasting glucose from <90 up to 99 mg/dL were associated with progressive impairments in surrogate markers of β-cell function and insulin sensitivity. Similar findings were observed for 2-h postload glucose concentrations, with impairments in β-cell function and insulin sensitivity as postload glucose levels raised from less than 120 mg/dL up to 139 mg/dL. Expectedly, the presence of IFG and/or IGT was characterized by further worsening of insulin resistance and β-cell function, reflected by a 75% reduction in the disposition index, a 50% reduction in early pancreatic response and the Matsuda index, and 60%–90% increases in HOMA-IR.

In conclusion, by employing these indices, it was possible to demonstrate that even gradual increments in fasting and postload glucose concentrations within what is considered as the “normal range,” were associated with, and most likely result from, combined impairments in β-cell function and insulin sensitivity. Our findings support previous observations indicating that fasting glucose concentrations above 95 mg/dL are already associated with increased risk for developing T2DM and cardiovascular disease.⁵

Metabolic syndrome is a well-known risk factor for the development of T2DM and cardiovascular disease.^6,7,21,22 Here, we have demonstrated that the ability of the pancreas to release insulin in response to an oral glucose load and the sensitivity of tissues to the action of insulin are increasingly impaired as the number of traits concurring in a subject increase. Compared to subjects with no traits, euglycemic subjects with one or two traits show impairments in early pancreatic response and decreases in insulin sensitivity. These findings suggest that the sole presence of traits or components of metabolic syndrome may mark alterations in glucose and insulin homeostasis, suggestive of an increased risk of developing T2DM.

The degree of impairment in insulin sensitivity and β-cell function observed in subjects with four or five traits was much greater than that observed in subjects with three traits, even though both groups of patients met criteria of metabolic syndrome. Such a finding, most likely results from the fact that glucose abnormalities contributed mainly as a fourth or fifth trait, whereas obesity, low HDL, and high triglycerides contributed mainly to the first three traits. These findings indicate that diagnosis of metabolic syndrome does not imply that all subjects are at the same risk. In fact, even within the metabolic syndrome, there is a continuum of risk with an increase in number and severity of traits.²³

NECP ATP III and IDF guidelines employed different criteria for diagnosis of metabolic syndrome. For example, NECP ATP III gives equal weight to each trait or component,¹⁶ whereas according to IDF guidelines, abdominal obesity is a required component of the syndrome.^17,18 Interestingly, in our study population, subjects with abdominal obesity as the sole trait and subjects with low HDL-C as a single trait or even combined with high triglycerides had different profiles. Increases in HOMA-IR and decreases in the Matsuda index and disposition characterized subjects with abdominal obesity as a single trait, whereas no significant impairments in early pancreatic response and insulin sensitivity were found when low HDL-C was present as a single trait or even when combined with high triglyceride levels. In fact, in a large sample of nondiabetic Korean subjects, baseline BMI, and fasting glucose were strongest predictors of incident T2DM.²⁴ Our results suggest that all traits of metabolic syndrome are not equally associated with a state of insulin resistance and/or impaired β-cell function. Our findings favor the use of IDF guidelines to define metabolic syndrome where obesity is a required trait for the diagnosis of metabolic syndrome. In addition, insulin resistance was the major alteration observed in subjects with abdominal obesity as a single trait, because no evidence of early impairments in β-cell function was observed in these subjects.

In summary, these results suggest that at early stages, abdominal obesity is associated with a state of insulin resistance in the absence of detectable defects in β-cell function. In nonobese subjects, low HDL-C and high triglycerides, even if considered as traits of metabolic syndrome, were not found to reflect or induce changes in β-cell function and/or insulin sensitivity. Findings in subjects with different loads of risk factors, as assessed by presence of traits of metabolic syndrome and by fasting and postload glucose concentrations, revealed that impairments in insulin secretion and sensitivity appear much before the development of metabolic syndrome, IFG and IGT. By employing surrogate markers of early pancreatic insulin response and of insulin sensitivity, it was possible to detect progressive deterioration of glucose homeostasis much before fasting or postload hyperglycemia increase to threshold guidelines values. Such measurements showed early alterations in glucose homeostasis in subjects with one and two traits, not meeting criteria of metabolic syndrome. However, several factors appear to interact, namely, the type and number of traits coexisting in a subject, for how long each trait was present, and the severity of each trait.³

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Unwin

, Shaw

, Zimmer

, Alberti

KGMM

. Impaired glucose tolerance and impaired fasting glycemia: The current status on definition and intervention. Diabet Med, 2002; 19:709–723.

The Expert Committee on the Diagnostic and Classification of Diabetes Mellitus. Report of the Expert Committee on the Diagnostic and Classification of Diabetes Mellitus. Diabetes Care, 2008; 31:562–567.

Abdul-Ghani

, Abdul-Ghani

, Ali

, Defronzo

. One-hour plasma glucose concentration and the metabolic syndrome identify subjects at high risk for future type 2 diabetes. Diabetes Care, 2008; 31:1650–1655.

Abdul-Ghani

, Lyssenko

, Tuomi

et al. Fasting versus postload plasma glucose concentration and the risk for future type 2 diabetes: Results from the Botnia Study. Diabetes Care, 2009; 32:281–286.

Malik

, Wong

, Franklin

et al. Impact of the metabolic syndrome on mortality from coronary heart disease, cardiovascular disease and all causes in United States adults. Circulation, 2004; 110:1245–1250.

Sattar

, Gaw

, Scherbakova

et al. Metabolic syndrome with and without C-reactive protein as predictor of coronary heart disease and diabetes in the West Scotland Coronary Prevention Study. Circulation, 2003; 108:414–419.

Hunt

, Resendez

, Williams

et al. National Cholesterol Education program versus World Health Organization metabolic syndrome in relation to all-cause and cardiovascular mortality in the San Antonio Heart Study. Circulation, 2004; 110:1251–1257.

Tennenbaum

, Fisman

. “The metabolic syndrome … is dead”: These reports are an exaggeration. Cardiovasc Diabetol, 2011; 10:11–14.

Lindstrom

, Tuomillehto

. The diabetes risk score: A practical tool to predict type 2 diabetes at risk. Diabetes Care, 2003; 6:725–731.

10.

DeFronzo

. Lilly lecture 1987: The triumvirate: Beta-cell, muscle liver: A collusion responsible for NIDDM. Diabetes, 1988; 37:667–687.

11.

Hanson

, Pratley

, Bogardus

et al. Evaluation of simple indices of insulin sensitivity and insulin secretion for use in epidemiologic studies. Am J Epidemiol, 2000; 151:190–198.

12.

Abdul-Ghani

, Matsuda

, Jani

et al. The relationship between fasting hyperglycemia and insulin secretion in subjects with normal or impaired glucose tolerance. Am J Physiol Endocrinol Metab, 2008; 295:401–406.

13.

Matsuda

, DeFronzo

. Insulin sensitivity indices obtained from oral glucose tolerance testing. Diabetes Care, 1999; 22:1462–1470.

14.

Abdul-Ghani

, Matsuda

, Balas

et al. Muscle and liver insulin resistance indexes derived from the oral glucose tolerance test. Diabetes Care, 2007; 30:89–94.

15.

De Fronzo

, Abdul-Ghani

. Oral disposition index predicts the development of future diabetes above and beyond fasting and 2-h glucose levels. Diabetes Care, 2009; 32:e86–e88.

16.

Utzschneider

, Prigeon

, Faulenbach

et al. Oral disposition index predicts the development of future diabetes above and beyond fasting and 2-h glucose levels. Diabetes Care, 2009; 32:335–341.

17.

Abdul-Ghani

, Williams

, DeFronzo

et al.

What is the best predictor of future type 2 diabetes?

Diabetes Care, 2007; 30:1544–1548.

18.

National Cholesterol Education Program. Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of the High Blood Cholesterol in Adults (ATP III) JAMA, 2001; 285:2486–2497.

19.

Alberti

, Zimmet

, Shaw

. IDF Epidemiology Task Force Consensus Group. The metabolic syndrome: a new worldwide definition. Lancet, 2005; 366:1059–1062.

20.

International Diabetes Federation. The IDF consensus worldwide definition of the metabolic syndrome. http://www.idf.org/webdata/docs/IDF_Meta_def.final.pdf.

21.

Forsen

, Lahti

, Nissen

, Taskinen

, Groop

. Cardiovascular morbidity and mortality associated with the metabolic syndrome. Diabetes Care, 2001; 24:683–689.

22.

Azevedo

, Bettencourt

, Almeida

, Santos

, Abreu-Lima

, Hense

H-S

, Baroos

. Increasing number of components of the metabolic syndrome and cardiac structural and functional abnormalities-cross-sectional study of the general population. BMC Cardiovasc Disord, 2007; 7:17–26.

23.

Cubeddu

, Hoffmann

. Metabolic syndrome: An all or none or a continuum load of risk. Metab Syndr Relat Disord, 2012; 10:14–19.

24.

Lee

, Kwon

, Rhee

et al. The effect of body mass index and fasting glucose on the relationship between blood pressure and incident diabetes mellitus: A 5-year follow-up study. Hypertens Res, 2011; 34:1093–1097.