Diabetes Mellitus: A Fundamental and Clinical Text
3rd Edition
© 2004 Lippincott Williams & Wilkins
←↑→
53
Metabolic Abnormalities in the Development of Type 2 Diabetes Mellitus
P. Antonio Tataranni
Clifton Bogardus
Diabetes mellitus (DM) is a heterogeneous group of metabolic disorders characterized by hyperglycemia (1,2). The major subclasses of DM include type 1 DM (formerly known as insulin-dependent DM), type 2 DM (formerly non–insulin-dependent DM), and gestational DM. In addition, a large number of other uncommon conditions, including specific genetic defects that affect insulin secretion and insulin action, can cause DM. Type 1 DM is due, in most cases, to autoimmune destruction of pancreatic β-cells, resulting in an absolute deficiency of insulin and a requirement for exogenously administered insulin. In this chapter, subjects who meet the glycemic criteria for DM, and do not have type 1 DM, gestational DM, or DM due to a known defect, such as a mutation in the insulin receptor (3), glucokinase (4), or one of the hepatic nuclear transcription factors (5,6,7,8), are considered to have type 2 DM. As is apparent from its current definition, the cause of type 2 DM is unknown. However, considerable progress has been made to define the metabolic characteristics of people who have, or later acquire, type 2 DM. Cross-sectional, prospective, and longitudinal data relevant to understanding the development of type 2 DM from normal glucose tolerance are reviewed. As is shown, many investigators have reported similar data from different populations around the world. The consistency of the data is indicative of the reliability and general applicability of the findings and the conclusions to be drawn from them.
Cross-Sectional Data
Obesity
It was a generally accepted clinical impression for centuries (see review in reference 9) that fatter people were more likely to have DM. Overwhelming evidence has accumulated to prove this clinical impression as accurate. The association of obesity with type 2 DM has been observed in comparisons of different populations and within populations. In ten different populations from divergent areas of the world, West and Kalbfleisch (10) observed a remarkably strong correlation between population prevalences of type 2 DM (what they called “maturity-onset” DM) and mean percentage of standard weight of the populations. Similar associations were found in several intrapopulation studies, beginning as early as 1921 in a study of the U.S. population by Joslin (11), and subsequently in other populations. Although the association of type 2 DM with obesity is remarkable in its consistency, a small percentage of people who are lean have type 2 DM (12,13). In these studies, subjects were classified as nonobese based on their body mass index (BMI, weight divided by height squared). BMI is an imprecise index of body composition, however, and in certain groups may underestimate the actual proportion of body fat as measured using direct methods (14). Thus, it is unknown whether these cases represent a distinct subtype of DM or simply reflect the normal variance present in human populations.
In addition to being more obese, on average, than nondiabetic people, people with type 2 DM have a more central distribution of body fat. As early as 1956, Vague (15) suggested that central obesity, or what he called “android obesity,” was more common in people with DM. Feldman et al. (16), using different measures of body fat distribution, confirmed these findings. In 1984 Hartz et al. (17), in a survey of over 30,000 American women, reported a higher prevalence of DM in women with a greater proportion of body fat at the waist relative to the hips even at comparable degrees of obesity. In addition to these intrapopulation surveys, West (9) reported that the Plains Indians of Oklahoma, a population with a high prevalence of type 2 DM, had a more central distribution of body fat compared with whites in the U.S. population, a group with a lower prevalence of type 2 DM.
Despite the large number of studies in which the degree of obesity of patients with type 2 DM has been compared with that of nondiabetic subjects, fewer data have been reported on the relationship between degree of obesity (and central obesity) and the extent of glycemia across a wide range of glycemia. These relationships in Pima Indians are shown in Fig. 53.1. It is apparent that on average those with type 2 DM are more obese, and more centrally obese, than nondiabetic subjects. In addition, the most hyperglycemic people with diabetes are less obese than the less severely diabetic individuals. Glycemia is negatively correlated with degree of obesity expressed as percentage
P.798

of body fat in men and women with type 2 DM. Glycemia is unrelated to the distribution of fat in diabetic people. In contrast to these correlations in diabetic subjects, in nondiabetic subjects, glycemia is positively correlated with percentage body fat and distribution of fat.
Figure 53.1. The relationship between percentage body fat (% Fat) and central distribution of fat (Waist/Thigh) and the plasma glucose concentration 2 hours after ingestion of 75 g of glucose in Pima Indians. Percentage fat was estimated by hydrodensitometry, and the waist/thigh ratio was measured as the ratio of the waist circumference at the umbilicus (supine) and the thigh circumference at the gluteal fold (standing).
Insulin Secretory Dysfunction
Definition of the relationship between glycemia and insulin secretory function began with the introduction of the insulin radioimmunoassay by Yalow and Berson (18). Reaven and Miller (19) were the first to report an inverted U–shaped relationship between insulin and glucose responses to glucose ingestion in lean and obese people. In their white subjects, those with moderately impaired glucose tolerance, or with type 2 DM and modest elevations of plasma glucose concentrations, had the highest insulin responses. Compared with this group, patients with type 2 DM and more severe hyperglycemia had lower insulin responses, but the responses were not lower than those with normal glucose tolerance unless the hyperglycemia was extreme. This inverted U–shaped relationship of insulin and glucose response to oral glucose has subsequently been confirmed by other investigators in other populations in Australia (20) and the United States (21).
Further clarification of the relationship between glycemia and insulin secretory function in humans came with the introduction of the intravenous glucose tolerance test. The insulin response to intravenously injected glucose was found to be biphasic, and people with type 2 DM, even with only mild fasting hyperglycemia or mild hyperglycemia after glucose ingestion, had a marked defect in first-phase response (22). This lack of a first-phase response, or acute insulin response (AIR), in people with type 2 DM has been observed by many investigators (12,22). Subsequent work indicated that a decreased AIR in people with type 2 DM was specific for glucose, as opposed to other islet cell secretagogues such as arginine (23). Halter et al. (24) pointed out, however, that the AIR depended on the prestimulus plasma glucose concentration. At similar prestimulus plasma glucose concentrations, people with type 2 DM had reduced AIRs in response to a variety of β-cell stimuli (24). From these studies, it was clear that all people with type 2 DM, regardless of a measurable hyperinsulinemic response to oral glucose, had a detectable defect in insulin secretory function. The physiologic relevance of a reduced AIR to an unphysiologic stimulus (intravenous glucose) has been questioned. However, it should be noted that AIR is related to more physiologic measures of insulin secretion. It was reported that the AIR is positively correlated with early insulin responses 30 minutes after ingestion of glucose (25). Results in nondiabetic Pima Indians
P.799

indicate further that the AIR is positively correlated with early insulin responses 30 minutes after ingesting a mixed meal (Table 53.1). Because AIR is also positively correlated with insulin action, this relationship could be a result of insulin resistance. This is not the case, however. AIR is positively correlated with the plasma insulin concentration 30 minutes after a meal, independent of insulin resistance (Table 53.1).
The cross-sectional relationships between AIR, fasting insulin concentration, insulin responses to oral glucose or a mixed meal, and glucose concentration 2 hours after the ingestion of 75 g of glucose in Pima Indians are summarized in Fig. 53.2. These data on approximately 400 Pimas confirm the results collected in various other groups and demonstrate two additional features of the relationship between glycemia and insulin secretory function in humans. Among those people with untreated type 2 DM, fasting insulin concentrations and the insulin responses to oral glucose and a mixed meal are negatively correlated with glycemia. In contrast to these negative relationships in people with type 2 DM, among nondiabetic subjects, fasting insulin concentrations and insulin responses to oral glucose and a mixed meal are positively correlated with glycemia.
The cross-sectional relationship between AIR and glycemia in Pimas is quite different from the relationships of insulin responses with glycemic responses to oral nutrients. As in other populations, the AIR to intravenous glucose is absent in many people with type 2 DM (21,22), including those with only slightly higher glucoses than nondiabetic people. An inverted U relationship between AIR and glycemia is found among the nondiabetic subjects as well, however. Thus, the inflection point beyond which AIR declines with increasing glycemia is substantially less than is seen with insulin responses to oral glucose or mixed meals.
Insulin Resistance
Based on his pioneering studies of glucose responses to combined glucose ingestion and intravenous insulin infusion, Himsworth (26) was the first to suggest, in the 1940s, that some people with DM were hyperglycemic because of resistance to the action of insulin to remove glucose from the blood rather than because of insulin deficiency. His work was apparently largely ignored until Yalow and Berson (27) reported in 1960 that some people with type 2 DM were hyperinsulinemic. In the ensuing decades, different investigators, using many different methods in studies of various populations, have confirmed that insulin resistance is a consistent feature of type 2 DM. Several groups of investigators (28,29,30) measured the rate of insulin-mediated glucose uptake of the forearm in subjects with and without type 2 DM. If their results were corrected for the effect of hyperglycemia to increase glucose uptake, it was evident that those with type 2 DM responded less well to insulin. Early evidence for insulin resistance of the whole body in people with type 2 DM was that the rate of decline of the plasma glucose concentration in response to insulin infusion was reduced in those with fasting hyperglycemia (31). To avoid hypoglycemia during such tests, and to quantify better the degree of insulin resistance, techniques were developed to evaluate whole-body insulin action in nondiabetic people and people with type 2 DM at similar and constant plasma insulin concentrations. Endogenous insulin secretion was inhibited using either a combination of epinephrine and propranolol (32) or somatostatin (33). The plasma glycemic response to a constant infusion of insulin and glucose was then a measure of insulin action in the whole body. Using such techniques, Reaven and colleagues (32) and, later, Harano et al. (33) found that people with type 2 DM were more insulin resistant than nondiabetic subjects.
Table 53.1. Predicting the plasma insulin concentration 30 minutes after a mixed meal in nondiabetic Pima Indians (N = 296) using multiple linear regression
Models and covariants Standardized betaa p value for beta Overall R2 Overall p value
1. AIRb 0.285 0.0001 0.081 0.0001
2. AIR 0.235 0.0001 — —
    Mc -0.360 0.0001 0.209 0.0001
aBetas were standardized by standardizing the covariates to a mean of zero and a standard deviation of 1.
bAIR, acute insulin response from 3 to 5 minutes after a bolus intravenous injection of 25 g of glucose.
cM, insulin-mediated glucose uptake rate at a plasma insulin concentration of ∼130 mU/ml, as determined by the hyperinsulinemic, euglycemic clamp technique (see ref. 34 for methodology).
In 1979, DeFronzo et al. (34) introduced another method to measure whole-body insulin action in vivo, the glucose/insulin clamp technique. With this technique, whole-body insulin action is quantified as the rate of a variable glucose infusion required to maintain a constant plasma glucose concentration during a constant insulin infusion. Compared with nondiabetic subjects, people with type 2 DM were found by many investigators to be insulin resistant (21,22,34,35,36,37,38).
More recently, Bergman et al. (39,40) introduced a technique in which an index of insulin sensitivity could be derived by applying mathematical modeling to data derived from an insulin- or tolbutamide-modified intravenous glucose tolerance test. Estimates of insulin sensitivity obtained using this method correlate well with those from the clamp technique (41). This approach has advantages over the glucose clamp in that it is easier to perform and is thus more amenable to large-scale epidemiologic studies. These studies have confirmed that most
P.800

people with type 2 DM are insulin resistant, although a small percentage of people with type 2 DM could be classified as insulin sensitive (42).
Figure 53.2. The relationships between fasting plasma insulin concentrations (A), the 3-hour insulin response to ingestion of 75 g of glucose (B), the 4-hour insulin response to ingestion of a standard mixed meal (C), and the acute insulin response to an intravenous glucose bolus (D), and the plasma glucose concentration 2 hours after ingestion of 75 g of glucose. The best-fit, significant regression between the two variables is shown, determined separately in nondiabetic and diabetic subjects.
The simple dichotomous studies of nondiabetic subjects compared with people with type 2 DM have been expanded to examine the relationship between glycemia and insulin action in vivo across the entire range of glycemia. The consistency of this relationship found by different investigators in both lean and obese groups is apparent from Fig. 53.3. As indicated in the figure, and as reported in studies of the Pima Indians (37), at physiologic plasma insulin concentrations there is little or no correlation between glycemia and insulin action in vivo in people with type 2 DM, whereas in nondiabetic people these metabolic variables are negatively correlated.
Many investigators have also determined which of the insulin-sensitive tissues (i.e., adipose tissue, skeletal muscle, and liver) are insulin resistant in people with type 2 DM. Decreased insulin action on insulin receptor signaling and glucose transport has been found in isolated adipocytes and skeletal muscle tissue obtained from whites (43,44,45,46,47,48,49,50) and Pima Indians with type 2 DM (51,52,53,54,55,56,57). While the complete chain of molecular steps leading from the interaction of insulin with its receptor to the altered biologic effects of insulin in people with type 2 DM remains to be fully elucidated (58,59), experiments in which the glucose/insulin clamp technique was combined with calorimetry or measures of arteriovenous differences in glucose across the leg (37,60,61) indicated indirectly that people with type 2 DM were especially resistant to the action of insulin on skeletal muscle glycogenesis. In 1989, Schulman et al. (62) demonstrated directly that the rate of insulin-mediated glycogenesis, as determined by nuclear magnetic resonance, was reduced in people with type 2 DM compared with nondiabetic subjects.
Caro et al. (43) obtained liver tissue by open biopsy during gastric bypass surgery in obese nondiabetic subjects and obese people with type 2 DM. They reported decreased insulin-stimulated insulin receptor tyrosine kinase activity and decreased insulin-stimulated aminoisobutyric acid uptake into isolated hepatocytes in those with DM compared with nondiabetic people. This in vitro evidence of hepatic insulin resistance in people with type 2 DM is consistent with results of clinical experiments. During a glucose/insulin clamp procedure, the rate of endogenous glucose production can be quantitated using glucose tracers. Many studies have confirmed that there is a rightward shift in the insulin dose-response curve for suppression of endogenous glucose production in people with type 2 DM compared with control subjects (36,37). At high plasma insulin concentrations, there is complete suppression of endogenous glucose production in diabetic and nondiabetic subjects. At lower, more physiologic plasma insulin concentrations, endogenous glucose production is fully suppressed in nondiabetic people, but incompletely suppressed in people with type 2 DM, indicating hepatic insulin resistance in the latter group.
Obesity undoubtedly is a contributing cause of insulin resistance in most people with type 2 DM. Data in whites and Pima Indians, however, indicate that it is likely not the only cause of insulin resistance (63,64,65). In both populations, insulin resistance is not linearly correlated with degree of obesity in nondiabetic
P.801

people. Although in leaner to moderately obese subjects obesity is negatively correlated with insulin action, above a percentage body fat of approximately 30% in Pima Indians, and above approximately 130% ideal body weight in whites, obesity and insulin action in vivo are unrelated. Equally important, at any given percentage body fat or BMI there is considerable variance in insulin action that is not attributable to obesity (65,66). As predicted from these relationships in nondiabetic subjects, in Pimas with type 2 DM, who are in general obese and insulin resistant, there is no negative correlation between degree of obesity and insulin action (data not shown).
Figure 53.3. The relationship between insulin action in vivo and fasting plasma glucose concentrations in three different studies. Insulin action was estimated in each study using the hyperinsulinemic, euglycemic clamp technique at physiologic plasma insulin concentrations. (A redrawn from
Unger RH. Lipotoxicity in the pathogenesis of obesity-dependent NIDDM: genetic and clinical implications. Diabetes 1995;44:863
, with permission; B redrawn from
Welborn TA, Stenhouse NS, Johnstone CG. Factors determining serum-insulin response in a population sample. Diabetologia 1969;5:263
, with permission; C redrawn from
Kolterman OG, Gray RS, Griffin J, et al. Receptor and post-receptor defects contribute to the insulin resistance in non-insulin-dependent diabetes mellitus. J Clin Invest 1981;68:957
, with permission.)
Excess Endogenous Glucose Production
Most investigators, using either glucose tracer methodology or organ balance techniques, have reported excess rates of postabsorptive endogenous glucose production in people with type 2 DM compared with nondiabetic subjects (21,36,37). It has also consistently been found in several white groups and in Pima Indians that the fasting plasma glucose concentration is positively correlated with the rate of fasting endogenous glucose production in people with type 2 DM, which in general is not true in nondiabetic subjects (21,36,37). A major action of insulin in the liver is to suppress endogenous glucose production. It might be expected, therefore, that people with DM with hepatic insulin resistance might also have an excess rate of postabsorptive endogenous glucose production, especially in those with fasting plasma insulin concentrations that are similar to, or below, those of nondiabetic subjects (see earlier section on Insulin Secretory Dysfunction). In addition to direct effects on the liver, insulin has indirect effects that result in the suppression of endogenous glucose production. Indeed, indirect effects, which are thought to involve the suppression of adipocyte lipolysis by insulin, resulting in lowered free fatty acid delivery to the liver, may be quantitatively more important than insulin’s direct effects on the liver (67).
Summary and Implications of Cross-Sectional Data
Compared with nondiabetic subjects, people with type 2 DM, on average, (a) are more obese, particularly centrally; (b) have abnormal insulin secretory function; (c) are insulin resistant in all three insulin-responsive tissues; and (d) have an excess rate of endogenous glucose production. Although some people with type 2 DM may be relatively lean or insulin sensitive, this is uncommon. The fact that these average metabolic characteristics of people with type 2 DM have been found so consistently by so many different investigators in divergent populations and are present in most patients to one degree or another indicates that that they should be widely accepted without controversy.
However, there is more to be inferred from these cross-sectional data than just a comparison of diabetic and nondiabetic groups. Within the diabetic group, the degree of glycemia is positively correlated with the rate of endogenous glucose production and is negatively correlated with insulin responses to glucose. Glycemia is not correlated with the degree of insulin resistance in people with DM, and the most severely hyperglycemic patients with DM are less obese than those who are less hyperglycemic. Thus, although people with type 2 DM are, on average, obese and insulin resistant, the metabolic characteristics that distinguish the more hyperglycemic from the less hyperglycemic patients with DM are reduced insulin secretion and excess endogenous glucose production.
P.802

The relationships between obesity, insulin secretion, insulin resistance, and endogenous glucose production are considerably different in the nondiabetic subjects. In this group, increasing glycemia is not correlated with fasting rates of endogenous glucose production, but is positively correlated with increasing obesity (and central obesity), increasing severity of insulin resistance, and increasing hyperinsulinemia in response to oral nutrients. There is an inverted U–shaped relationship between the AIR to intravenous glucose and glycemia in this group, similar to the relationship of insulin responses to oral nutrients with glycemia in the entire population of diabetic and nondiabetic subjects.
First, it can be inferred from these data in nondiabetic subjects that insulin resistance, due entirely or partly to obesity and central obesity, is a contributing cause of higher glucose concentrations after glucose ingestion in this group. The contribution of abnormal secretory function to hyperglycemia in the nondiabetic subjects is less apparent. The most hyperglycemic nondiabetic people have the highest insulin responses to ingestion of glucose and a mixed meal. Conversely, the same people have lower AIRs to an intravenous glucose stimulus compared with some nondiabetic subjects who are less hyperglycemic. Because the AIR significantly predicts early insulin responses to a mixed meal, it can be inferred that if the AIR of the most hyperglycemic nondiabetic subjects was as high as that found in some of the less hyperglycemic nondiabetic subjects, the early insulin response to a mixed meal would be even greater and the degree of hyperglycemia less. Thus, both insulin resistance and abnormal insulin secretory function contribute to hyperglycemia in the most hyperglycemic nondiabetic subjects. Compared with the most glucose-tolerant subjects, nondiabetic subjects with modest increases in glycemia have both higher insulin responses to ingested nutrients and higher AIRs. The modest increase in glycemia in these people, therefore, can be attributed to insulin resistance alone. Thus, variance in both insulin action and insulin secretory function, in varying degrees, account for the variance in glucose tolerance in nondiabetic subjects.
Prospective Data
Large numbers of cross-sectional studies have firmly established the metabolic characteristics of people with type 2 DM. Several characteristics of people with type 2 DM are observed in nondiabetic subjects as well, but it cannot be determined from cross-sectional studies alone which characteristic is a prediabetic abnormality. Such conclusions can be drawn only from prospective studies in which nondiabetic subjects are metabolically characterized and followed over time to determine who does and does not acquire type 2 DM. A major advance has been the large number of prospective studies that have established which metabolic abnormalities are prediabetic and which are not.
Obesity
Prospective studies of whites in Norway (68), Sweden (69), Israel (70,71), and the United States (72,73,74), as well as in Mexican Americans in Texas (75) and in Pima Indians in Arizona (76), have shown conclusively that obesity is a major risk factor for type 2 DM. In addition, studies of Israelis and Pima Indians have indicated that duration of obesity, in addition to degree of obesity, is a risk factor for type 2 DM (71,77). The risk for type 2 DM in Pima Indians is twice as high in those who have been obese for 10 years or more compared with those obese for less than 5 years (77).
A central distribution of body fat is also a major risk factor for type 2 DM, in addition to the effect of degree of obesity per se. In Swedish men matched for BMI, a central distribution of fat, as estimated by the ratio of waist-to-hip circumferences, is a major risk factor for type 2 DM (78); and in Pima Indians, central obesity, as estimated by the ratio of waist-to-thigh circumferences, is an additional risk factor for the disease (79). In Japanese-American men, increased intraabdominal fat, estimated using computed tomography, increases the risk for type 2 DM (80).
It is clear, however, that obesity is not the only major risk factor for type 2 DM. Although many U.S. whites are obese, less than 10% of the population has type 2 DM. Among whites (72) as well as in Pima Indians (76), parental DM is a major risk factor for type 2 DM, independent of the effect of obesity. This parental effect on the risk for type 2 DM must be due to other familial, probably genetic, metabolic abnormalities that affect insulin secretory function, insulin resistance, or endogenous glucose production.
Insulin Secretory Dysfunction
Four groups of investigators have reported the results of prospective studies in which early insulin secretory responses to intravenous glucose were measured in nondiabetic subjects and the subjects followed for several years to determine who acquired type 2 DM. These studies included nondiabetic subjects who had normal glucose tolerance [by criteria of the World Health Organization (2)], as well as those who had abnormal or impaired glucose tolerance. Based on studies of a Swedish population, Efendic et al. (81) concluded that “subjects with an inappropriately low insulin response in relation to insulin sensitivity” were prediabetic. Similarly, prospective studies of nondiabetic Pima Indians indicated that a relatively low AIR to intravenous glucose was a risk factor for type 2 DM, but only when referenced to the degree of insulin resistance (79). Skarfors et al. (82) found that a low incremental insulin response to intravenous glucose predicted DM independently of obesity and the fasting insulin concentration in 50-year-old Swedish men who were followed for an average of 10 years. Finally, Lundgren et al. (83) found a low AIR to be a predictor of type 2 DM in Swedish women. They observed further that many women with a low response did not acquire DM, and a few women with the highest insulin response did acquire DM; that is, “no safe level could be identified above which diabetes did not appear.” This is similar to the results found among the Pimas, where DM developed in some people with the highest insulin responses (79).
In contrast to the four studies of subjects with normal and impaired glucose tolerance, Warram et al. (84)
P.803

and Martin et al. (85) prospectively studied offspring of two parents with type 2 DM who had normal glucose tolerance at baseline examination. In their analyses, a low AIR was not predictive of type 2 DM. This is different from the results in Pima Indians, where a low AIR was a predictor of type 2 DM if the analyses were restricted to those who initially had normal glucose tolerance (79).
There are several prospective studies of the predictive effect of insulin responses to glucose ingestion and of the late, or second-phase, insulin response to intravenous glucose. These can in general be divided into studies of subjects who originally had normal glucose tolerance or impaired glucose tolerance. In subjects with impaired glucose tolerance, a low insulin response 2 hours after glucose ingestion is predictive of type 2 DM in French men (86), Nauruans (87), Japanese (88), and Pima Indians (89). In those with normal glucose tolerance, the converse has been found; namely, a high, late insulin response is predictive of type 2 DM in normal, glucose-tolerant U.S. whites (84,85), Nauruans (87), and Pima Indians (89).
Insulin Resistance
The fasting plasma insulin concentration is well correlated with the degree of insulin resistance in nondiabetic subjects and therefore has been used as a surrogate measure of whole-body insulin action in many prospective studies. A high fasting plasma insulin concentration has been found to be a major risk factor for type 2 DM in people with either normal glucose tolerance or impaired glucose tolerance, or both, in several populations, including whites in the United States and Sweden (81,82,83,84,85), French men (86), Mexican Americans (75), Nauruans (87), and Pima Indians (89).
Confirmation of these findings implicating insulin resistance as a major risk factor for type 2 DM has been reported in whites (84,85) and Pima Indians (79), where more accurate measures of insulin action in the whole body were made. In the study of whites (84,85), intravenous glucose tolerance tests were performed in the normal glucose-tolerant offspring of two parents with type 2 DM, and the offspring were followed for approximately 15 years to determine DM status. Insulin action was estimated from the results of the intravenous glucose tolerance test using either the Kg (the rate of decline of the plasma glucose concentration) (84) or by using the Bergman minimal model (85). The most insulin-resistant subjects in this study had a severalfold increased risk for type 2 DM (Fig. 53.4). In studies of Pima Indians, insulin action in vivo was measured using the hyperinsulinemic euglycemic clamp technique. Compared with the most insulin-sensitive Pimas (90th percentile), the most insulin-resistant Pimas (10th percentile) had a much greater risk for type 2 DM after 6 years of follow-up (79) (Fig. 53.5). In both these studies, insulin resistance was a major risk factor for type 2 DM in addition to the effect of obesity, and in the Pima study, in addition to the effect of central obesity.
Figure 53.4. Results of a prospective study of 155 offspring of parents with type 2 diabetes mellitus (DM) attending the Joslin Clinic. □, remained nondiabetic (n = 126); ▪, acquired DM (n = 25); SI, insulin sensitivity as determined from the intravenous glucose tolerance test using the Bergman minimal model. (Redrawn from
Warram JH, Martin BC, Krolewski AS, et al. Slow glucose removal rate and hyperinsulinemia precede the development of type II diabetes in the offspring of diabetic parents. Ann Intern Med 1990;113:909
, with permission.)
Excess Endogenous Glucose Production
The rate of postabsorptive endogenous glucose production has been measured only in a prospective study of Pima Indians (79). In that study, the rate of endogenous glucose production, measured after an overnight fast using glucose tracers, was not predictive of type 2 DM. However, the extent of insulin-mediated suppression of endogenous glucose production measured during the hyperinsulinemic clamp was a weak predictor of type 2 DM and was indicative of hepatic insulin resistance as a weak risk factor for the disease (79).
Summary and Implications of Prospective Data
Obesity is a major risk factor for type 2 DM, and a central distribution of fat adds additional risk. Insulin resistance is another,
P.804

independent, major risk factor. These data have been confirmed by different investigators in different populations and are found in nondiabetic subjects with normal glucose tolerance or impaired glucose tolerance. There are no reports in which obesity and insulin resistance were found not to be predictive of type 2 DM.
Figure 53.5. Proportional hazard analysis indicating the 8-year cumulative incidence rate of type 2 diabetes mellitus by tertiles of acute insulin response to glucose (AIR) and insulin action at physiologic plasma insulin concentrations (M) in 317 Pima Indians with normal glucose tolerance, 62 of whom developed diabetes over an average duration of follow-up of about 7 years. AIR and M were both predictive of type 2 DM (both p < 0.01) independent of age, sex, and degree of obesity.
Based on data from different populations with impaired glucose tolerance, it is also firmly established that reduced insulin secretion hours after glucose ingestion, or 10 to 60 minutes after intravenous glucose infusion, is predictive of type 2 DM. Conversely, in those with normal glucose tolerance, a late hyperinsulinemic response is predictive. In addition, among normal glucose-tolerant subjects, a lower AIR is frequently, but not universally, predictive of type 2 DM.
Excess fasting endogenous glucose production was not predictive of type 2 DM in the one study in which it was measured. Although this needs to be repeated in other populations, the result is consistent with the cross-sectional observation that the rate of fasting endogenous glucose production is unrelated to the degree of glucose tolerance among nondiabetic subjects.
Of the four metabolic characteristics of people with type 2 DM—obesity, insulin resistance, insulin secretory dysfunction, and excess endogenous glucose production—all but the last are found among nondiabetic subjects with some impairment of glucose tolerance, and each abnormality increases the risk for type 2 DM. Among nondiabetic subjects with normal glucose tolerance, obesity and insulin resistance are major risk factors, and a relatively lower AIR appears to be an additional risk factor.
Longitudinal Data
Cross-sectional studies and prospective studies have not fully resolved the relative importance of insulin resistance and insulin secretory dysfunction in the evolution of abnormal glucose tolerance. This can be addressed in a longitudinal study, in which insulin resistance, insulin secretion, and endogenous glucose production are measured in subjects at each stage in the progression from normal to impaired glucose tolerance to DM (90). Such a study, feasible only in populations with a high incidence of DM, was initiated in 1982 in the Pima Indians of Arizona. Longitudinal data were collected on 17 people (progressors) in whom glucose tolerance deteriorated from normal to impaired to diabetic over an average of 5 years. Data from a control group of 31 individuals who remained normally glucose tolerant over the same period also were analyzed (nonprogressors). Mean body weight increased by approximately 13 kg in the progressors and by approximately 6 kg in the nonprogressors. Commensurate with the degree of obesity, insulin action decreased in both groups. However, changes in AIR were very different between the two groups. AIR decreased by 27% with the transition from normal to impaired glucose tolerance and declined further (51%) with the development of diabetes. In contrast, AIR remained unchanged or increased slightly in the nonprogressors (Fig. 53.6). The rate of basal endogenous glucose production remained unchanged in the nonprogressors over the period of follow-up. In the progressors, the rate of basal endogenous glucose production remained unchanged with the transition from normal to impaired glucose tolerance and increased slightly with the development of diabetes (90). These results indicate that defects in both insulin secretion and insulin action contribute to the decline in glucose tolerance as people transition from normal to impaired glucose tolerance and, finally, to DM. Increases in endogenous glucose output, in contrast, occur only during the transition from impaired glucose tolerance to DM.
Figure 53.6. Relationship between the acute insulin response to glucose (AIR) and insulin action, measured at physiologic concentrations of insulin (M). The dashed lines represent the 95% confidence limits of the regression line (solid line) between these variables in persons with normal glucose tolerance. Also shown are the mean changes in these variables in those who developed diabetes (progressors) and in a control group who remained normal glucose tolerant over the same 5-year period (nonprogressors). EMBS, estimated metabolic body size. (Adapted from
Weyer C, Bogardus C, Mott DM, Pratley RE. The natural history of insulin secretory dysfunction and insulin resistance in the pathogenesis of type 2 diabetes mellitus. J Clin Invest 1999;104:787–794
, with permission.)
Overall Summary and Conclusions
It is clear from cross-sectional data collected from many different populations that, compared with nondiabetic subjects with normal glucose tolerance, nondiabetic subjects with impaired glucose tolerance are, on average, more obese, more insulin resistant, and more hyperinsulinemic in response to oral nutrients. However, they also have lower AIRs to intravenous glucose, and the physiologic relevance of this is attested to by its correlation with early insulin responses after a mixed meal. Thus, except for excess fasting endogenous glucose production, nondiabetic subjects with impaired glucose tolerance have all the abnormal characteristics of people with type 2 DM, although differing in severity and proportion. It is not surprising, therefore, that each of these metabolic defects—obesity, insulin resistance, and insulin secretory dysfunction—has been uniformly found to be predictive of type 2 DM in nondiabetic subjects with impaired glucose tolerance.
The available prospective data have also uniformly identified obesity and insulin resistance as major risk factors for type 2 DM among nondiabetic subjects with normal
P.805

glucose tolerance. In some, but not all studies, abnormal insulin secretory function was also predictive of type 2 DM in people with normal glucose tolerance. Thus, insulin resistance and insulin secretory dysfunction are metabolic abnormalities that can be identified in prediabetic subjects years before they acquire type 2 DM. These abnormalities worsen as glucose tolerance deteriorates from normal to impaired glucose tolerance and, finally, to DM. Increases in endogenous glucose production are evident only after the onset of DM and worsen in proportion to the severity of fasting hyperglycemia.
If these data are considered in the context of DM due to other causes, a generalized view of the risk for DM in various populations emerges. Assuming obesity affects DM risk only by affecting insulin action or insulin secretion, then the risk for DM can be viewed as a function of variations in insulin action and insulin secretion. In this scheme, several points are well established. When insulin secretion is reduced to zero as a result of destruction of the β-cells by autoimmune mechanisms (e.g., in type 1 DM or by chronic pancreatitis), the lifetime risk for DM approaches 100%. In people with the most severe degrees of insulin resistance (e.g., insulin receptor mutations), DM risk is extremely high (91), and only those with the largest insulin secretory responses do not acquire the disease. Between these extremes lie the more common situations. The risk for DM increases approximately linearly, and only gradually, from high to low insulin secretory function until the lowest levels of insulin secretion, when DM risk increases more abruptly. The slope of the relationship between DM risk and insulin secretory function becomes steeper in those who are more insulin resistant (i.e., the risk for DM increases geometrically relative to decreases in insulin action).
Relative to U.S. whites, Pima Indians have larger insulin secretory responses (92), but they have an increased risk for type 2 DM because of greater degrees of insulin resistance (92). The Mexican-American population has a degree of risk that lies between these two groups, and people with glucokinase or hepatic nuclear transcription factor mutations (4,5,6,7,8) have an increased risk for DM due to a defect in insulin secretion. There is, of course, considerable overlap between these groups. For example, there are likely some whites who are as insulin resistant as the average Pima Indian, but because their insulin secretory function is not as high, they may actually have a greater risk for type 2 DM than most Pimas. Similarly, there are Pimas with insulin secretory function similar to that of the average white, which, due to their greater insulin resistance, would increase the risk for type 2 DM.
Although prospective and longitudinal data have clarified the relative roles of insulin resistance and insulin secretory dysfunction in the evolution of abnormal glucose tolerance, it remains unclear how these physiologic factors interact to result in type 2 DM. It is easy to understand how DM develops in the absence of insulin secretion, but how does insulin resistance gradually result in the disease? Does insulin secretory function simply decrease with age? Chronic mild hyperglycemia gradually impairs β-cell function (93), and it has been suggested that hyperglycemia can worsen insulin resistance (21). The precise mechanisms of this “glucose toxicity” effect are unknown, however. It is also not known how a relatively lower AIR may increase the risk of type 2 DM. Because a lower AIR may be associated with higher plasma glucose concentrations within 30 minutes after a meal, could its effect on DM risk also be due to a glucose toxicity effect? The concept of “lipotoxicity” has been advanced to explain the pathogenesis of type 2 DM (94). The proponents of this theory propose that sustained increases in plasma and intracellular fatty acids lead to the progressive abnormalities in insulin action and insulin secretion that characterize the development of type 2 DM. Indeed, experimental data indicate that infusions of fatty acids lead to insulin resistance (95) and increase the basal insulin secretion (96), but decrease the AIR (97). More recently it has been observed that insulin resistance and type 2 DM are associated with a chronic activation of the innate immune system, manifested by abnormal circulating levels of inflammatory mediators (98,99,100,101,102,103). Proinflammatory cytokines or the acute-phase reactants that they stimulate are thought to impair insulin action or secretion. The origin of the association between inflammation and type 2 DM remains unknown. However, because the adipose tissue can produce a number of cytokines (tumor necrosis factor-α, interleukin-6, adiponectin), it is possible that the immune system mediates the effect of overnutrition on insulin resistance, insulin secretory dysfunction, and the later development of type 2 DM.
No simple metabolic defect is likely to explain the cause of type 2 DM in large numbers of people. A complete understanding of the causes of type 2 DM will require a better knowledge of the environmental and molecular genetic determinants of both insulin action and insulin secretory function, and, equally important, a better knowledge of how they interact pathophysiologically over time.
References
1. The 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(suppl 1):5.
2. World Health Organization Study Group. Diabetes mellitus. Technical reports series no. 727. Geneva: World Health Organization, 1985:9.
3. Taylor SI. Lilly Lecture. Molecular mechanisms of insulin resistance: lessons from patients with mutations in the insulin-receptor gene. Diabetes 1992;41:1473.
4. Froguelle P, Zouali H, Bionnet N, et al. Familial hyperglycemia due to mutations in glucokinase: definition of a new subtype of non–insulin-dependent (type 2) diabetes mellitus. N Engl J Med 1993; 328:697.
5. Yamagata K, Oda N, Kaisaki PJ, et al. Mutations in the hepatocyte nuclear factor-1-alpha gene in maturity onset diabetes of the young (MODY3). Nature 1996;384:455.
6. Yamagata K, Furata H, Oda N, et al. Mutations in the hepatocyte nuclear factor-4-alpha gene in maturity onset diabetes of the young (MODY1). Nature 1996;384:458.
7. Stoffers DA, Ferrer J, Clarke WL, et al. Early-onset type-II diabetes mellitus (MODY4) linked to IPF1. Nat Genet 1997;17:138.
8. Horikawa Y, Iwasaki N, Hara M, et al. Mutation in hepatocyte nuclear factor-1b gene (TCF2) associated with MODY. Nat Genet 1997;17:384–385.
9. West KM. Epidemiology of diabetes and its vascular lesions. New York: Elsevier, 1978.
10. West KM, Kalbfleisch JM. Influence of nutritional factors on prevalence of diabetes. Diabetes 1971;20:99.
11. Joslin EP. The prevention of diabetes mellitus. JAMA 1921;76:79.
P.806

12. Arner P, Pollare T, Lithell H. Different aetiologies of type 2 (non-insulin-dependent) diabetes mellitus in obese and non-obese subjects. Diabetologia 1991;34:483.
13. Mohan, V, Vijayaprabha R, Rema M, et al. Clinical profile of lean NIDDM in South India. Diabetes Res Clin Pract 1997;38:101.
14. Banerji MA, Faridi N, Atluri R, et al. Body composition, visceral fat, leptin, and insulin resistance in Asian Indian men. J Clin Endocrinol Metab 1999;84:137.
15. Vague J. The degree of masculine differentiation of obesities: a factor determining predisposition to diabetes, atherosclerosis, gout, and uric calculus disease. Am J Clin Nutr 1956;4:20.
16. Feldman R, Sender AJ, Siegelaub AB. Difference in diabetic and nondiabetic fat distribution patterns by skinfold measurements. Diabetes 1969;18:478.
17. Hartz AJ, Rupley DC, Rimm AA. The association of girth measurements with disease in 32,856 women. Am J Epidemiol 1984;119:71.
18. Yalow RS, Berson SA. Immunoassay of endogenous plasma insulin in man. J Clin Invest 1960;39:1157.
19. Reaven G, Miller R. Study of the relationship between glucose and insulin responses to an oral glucose load in man. Diabetes 1968;17: 560.
20. Welborn TA, Stenhouse NS, Johnstone CG. Factors determining serum-insulin response in a population sample. Diabetologia 1969; 5:263.
21. DeFronzo RA. Lilly Lecture 1987. The triumvirate: β-cell, muscle, liver: a collusion responsible for type 2 diabetes. Diabetes 1988;37:667.
22. DeFronzo RA, Ferrannini E. The pathogenesis of non–insulin-dependent diabetes: an update. Medicine (Baltimore) 1982;62:125.
23. Porte D Jr. Banting Lecture 1990. β-cells in type II diabetes mellitus. Diabetes 1991;40:166.
24. Halter JB, Graf RJ, Porte D Jr. Potentiation of insulin secretory responses by plasma glucose levels in man: evidence that hyperglycemia in diabetes compensates for impaired glucose potentiation. J Clin Endocrinol Metab 1979;48:946.
25. Yoneda H, Ikagami H, Yamamoto Y, et al. Analysis of early phase insulin responses in non-obese subjects with mild glucose intolerance. Diabetes Care 1992;15:1517.
26. Himsworth HP, Kerr RB. Insulin-sensitive and insulin-insensitive types of diabetes mellitus. Clin Sci 1942;4:120.
27. Yalow RS, Berson SA. Plasma insulin concentrations in nondiabetics and early diabetic subjects. Diabetes 1960;9:254.
28. Butterfield WJH, Whichelow MJ. Peripheral glucose metabolism in control subjects and diabetic patients during glucose, glucose-insulin and insulin sensitivity tests. Diabetologia 1965;1:43.
29. Jackson RA, Perry G, Rogers J, et al. Relationship between the basal glucose concentrations, glucose tolerance and forearm glucose uptake in maturity-onset diabetes. Diabetes 1973;22:751.
30. Zierler KI, Rabinowitz D. Roles of insulin and growth hormone, based on studies of forearm metabolism in man. Medicine (Baltimore) 1963;42:385.
31. Alford FP, Martin FIP, Pearson MJ. The significance of interpretation of mildly abnormal oral glucose tolerance. Diabetologia 1971;7:173.
32. Reaven GM, Farquhar JW. Steady state plasma insulin response to continuous glucose infusion in normal and diabetic subjects. Diabetes 1969;18:273.
33. Harano Y, Ohgaku S, Hidaka H, et al. Glucose insulin and somatostatin infusion for the determination of insulin sensitivity. J Clin Endocrinol Metab 1977;45:1124.
34. DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 1979;6:E214.
35. Reaven GM. Role of insulin resistance in human disease. Diabetes 1988;37:1595.
36. Kolterman OG, Gray RS, Griffin J, et al. Receptor and post-receptor defects contribute to the insulin resistance in non-insulin dependent diabetes mellitus. J Clin Invest 1981;68:957.
37. Bogardus C, Lillioja S, Howard B, et al. Relationships between insulin secretion, insulin action and fasting plasma glucose concentration in non-diabetic and non–insulin-dependent diabetic subjects. J Clin Invest 1984;74:1238.
38. Rizza RA, Mandarino LJ, Gerich JE. Mechanism and significance of insulin resistance in non–insulin-dependent diabetes mellitus. Diabetes 1981;30:990.
39. Bergman RN, Phillips LS, Cobelli C. Physiologic evaluation of factors controlling glucose tolerance in man: measurement of insulin sensitivity and beta-cell glucose sensitivity from the response to intravenous glucose. J Clin Invest 1981;68:1456.
40. Bergman RN. Lilly Lecture 1989. Toward physiological understanding of glucose tolerance: minimal-model approach. Diabetes 1989; 38:1512.
41. Saad MF, Anderson RL, Laws A, et al. A comparison between the minimal model and the glucose clamp in the assessment of insulin sensitivity across the spectrum of glucose tolerance: Insulin Resistance Atherosclerosis Study. Diabetes 1994;43:1114.
42. Haffner SM, Howard G, Mayer E, et al. Insulin sensitivity and acute insulin response in African-Americans, non-Hispanic whites, and Hispanics with NIDDM: the Insulin Resistance Atherosclerosis Study. Diabetes 1997;46:63.
43. Caro JF, Dohm LG, Pories WJ, et al. Cellular alterations in liver, skeletal muscle, and adipose tissue responsible for insulin resistance in obesity and type II diabetes mellitus. Diabetes Metab Rev 1989;5: 665.
44. Olefsky JM, Garvey WT, Henry RR, et al. Cellular mechanisms of insulin resistance in non–insulin-dependent (type II) diabetes. Am J Med 1988;85(suppl 5A):86.
45. Arner P, Pollare T, Lithell H, et al. Defective insulin receptor tyrosine kinase in human skeletal muscle in obesity and type 2 (non–insulin-dependent) diabetes mellitus. Diabetologia 1987;30:437.
46. Obermaier-Kusser B, White MF, Pongratz DE, et al. A defective intramolecular autoactivation cascade may cause the reduced kinase activity of the skeletal muscle insulin receptor from patients with non-insulin dependent diabetes mellitus. J Biol Chem 1989;264:9497.
47. Thorburn AW, Gumbiner B, Bulacan F, et al. Multiple defects in glycogen synthase activity contribute to reduced glycogen synthesis in non-insulin dependent diabetes mellitus. J Clin Invest 1991;87:489.
48. Damsbo P, Vaag A, Hother-Nielson O, et al. Reduced glycogen synthase activity in skeletal muscle from obese patients with and without type 2 (non–insulin-dependent) diabetes mellitus. Diabetologia 1991;34:239.
49. Wright KS, Beck-Nielsen H, Kolterman O, et al. Decreased activation of skeletal muscle glycogen synthase by mixed-meal ingestion in type 2 diabetes. Diabetes 1988;37:436.
50. Hansen T, Andersen CB, Echwald SM, et al. Identification of a common amino acid polymorphism in the p85alpha regulatory subunit of phosphatidylinositol 3-kinase: effects on glucose disappearance constant, glucose effectiveness, and the insulin sensitivity index. Diabetes 1997;46:494.
51. Foley JE. Mechanism of impaired insulin action in isolated adipocytes from obese and diabetic subjects. Diabetes Metab Rev 1988;4:487.
52. Bogardus C, Lillioja S, Stone K, et al. Correlation between muscle glycogen synthase activity and in vivo insulin action in man. J Clin Invest 1984;73:1185.
53. Kida YA, Esposito-Del Puente A, Bogardus C, et al. Insulin resistance is associated with reduced fasting and insulin-stimulated glycogen synthase phosphatase activity in human skeletal muscle. J Clin Invest 1990;85:476.
54. Nyomba BL, Ossowski VM, Bogardus C, et al. Insulin-sensitive tyrosine kinase: relationship with in vivo insulin action in human. Am J Physiol 1990;250:E964.
55. Kida Y, Nyomba BL, Bogardus C, Mott DM. Defective insulin response of cyclic adenosine monophosphate dependent protein kinase in insulin-resistant human. J Clin Invest 1991;87:673.
56. Sommercorn J, Fields R, Raz I, et al. Abnormal regulation of ribosomal protein S6 kinase in insulin skeletal muscle of insulin-resistant humans. J Clin Invest 1993;91:509.
57. McGuire MC, Fields RM, Nyomba BL, et al. Abnormal regulation of protein tyrosine phosphatase activities in skeletal muscle of insulin-resistant humans. Diabetes 1991;40:939.
58. Pessin JE, Saltiel AR. Signaling pathways in insulin action: molecular targets of insulin resistance. J Clin Invest 2000;106:165–169.
P.807

59. Kido Y, Nakae J, Accili D. Clinical review 125: the insulin receptor and its cellular targets. J Clin Endocrinol Metab 2001;86:972–979.
60. DeFronzo RA, Gunnarsson R, Bjorkman O, et al. Effects of insulin on peripheral and splanchnic glucose metabolism in noninsulin dependent (type II) diabetes mellitus. J Clin Invest 1985;76:149.
61. Kelly DE, Mokaw M, Mandarino LJ. Intracellular defects in glucose metabolism in obese patients with type 2 diabetes. Diabetes 1992;41: 698.
62. Shulman GI, Rothman DL, Jue T, et al. Quantitation of muscle glycogen synthesis in normal subjects and subjects with non–insulin-dependent diabetes by 13C nuclear magnetic resonance spectroscopy. N Engl J Med 1989;322:223.
63. Ferrannini E, Groop LC. Hepatic glucose production in insulin-resistant states. Diabetes Metab Rev 1989;5711.
64. Elbein SC, Ward WK, Bearch JC, et al. Familial type 2 diabetes: molecular-genetic analysis and assessment of insulin action and pancreatic β-cell function. Diabetes 1988;37:377.
65. Lillioja S, Bogardus C. Obesity and insulin resistance: lessons learned from the Pima Indians. Diabetes Metab Rev 1988;4:517.
66. Lillioja S, Mott DM, Zawadzki JK, et al. In vivo insulin action is a familial characteristic in non-diabetic Pima Indians. Diabetes 1987; 36:1329.
67. Mittelman SD, Fu YY, Rebrin K, et al. Indirect effect of insulin to suppress endogenous glucose production is dominant, even with hyperglucagonemia. J Clin Invest 1997;100:3121.
68. Westlund K, Nicolaysen R. Ten year mortality and morbidity related to serum cholesterol: a follow-up of 3,751 men aged 40–49. Scand J Lab Clin Invest 1972;30(suppl 127):1.
69. Ohlson L-O, Larsson B, Eriksson H, et al. Diabetes in Swedish middle-aged men: the study of men born in 1913 and 1923. Diabetologia 1987;30:386.
70. Madelie JH, Herman JB, Goldbourt U, et al. Variations in incidence of diabetes among 10,000 adult Israeli males and the factors related to their development. In: Levine R, Luft R, eds. Advances in metabolic disorders. New York: Academic, 1978:93.
71. Modan M, Karasik A, Halkin H, et al. Effect of past and concurrent body mass index on prevalence of glucose intolerance and type 2 (non–insulin-dependent) diabetes and on insulin response: the Israeli Study of Glucose Intolerance, Obesity and Hypertension. Diabetologia 1986;29:82.
72. O’Sullivan JB, Mahan CM. Blood sugar levels, glycosuria, and body weight related to development of diabetes mellitus. JAMA 1965; 194:587.
73. Dunn JP, Ipsen J, Elsom KO, et al. Risk factors in coronary artery disease, hypertension and diabetes. Am J Med Sci 1970;259:309.
74. Wilson PW, McGee DL, Kannel WB. Obesity, very low density lipoproteins, and glucose intolerance over fourteen years: the Framingham Study. Am J Epidemiol 1981;114:697.
75. Haffner SM, Stern MP, Mitchell BD, et al. Incidence of type II diabetes in Mexican-Americans predicted by fasting insulin and glucose levels, obesity and body fat distribution. Diabetes 1990;39:283.
76. Knowler WC, Pettitt DJ, Saad MF, et al. Diabetes mellitus in the Pima Indians: incidence risk factors and pathogenesis. Diabetes Metab Rev 1990;6:1.
77. Everhart JE, Pettitt DJ, Slaine KR, et al. Duration of obesity is a risk factor for non–insulin-dependent diabetes mellitus. Diabetes 1992; 41:235.
78. Ohlson L-D, Larsson B, Svärdsudd K, et al. The influence of body fat distribution on the incidence of diabetes mellitus: 13.5 years of follow-up of the participants in the study of men born in 1913. Diabetes 1985;34:1055.
79. Lillioja S, Mott DM, Spraul M, et al. Insulin resistance and insulin secretory dysfunction as precursors of non–insulin-dependent diabetes mellitus: prospective studies of Pima Indians. N Engl J Med 1993;329:1988.
80. Bergstrom RW, Newell-Morris LL, Leonetti DL, et al. Association of elevated fasting C-peptide levels and increased abdominal fat distribution with development of type 2 diabetes in Japanese-American men. Diabetes 1990;39:104.
81. Efendic S, Luft R, Wajngot A. Aspects of the pathogenesis of type 2 diabetes. Endocr Rev 1984;5:395.
82. Skarfors ET, Selinus KI, Lithell HO. Risk factors for developing non-insulin dependent diabetes: a 10 year follow up of men in Uppsala. BMJ 1991;303:755.
83. Lundgren H, Bengtsson C, Blohm G, et al. Fasting serum insulin concentration and early insulin response as risk determinants for developing diabetes. Diabetes Med 1990;7:407.
84. Warram JH, Martin BC, Krolewski AS, et al. Slow glucose removal rate and hyperinsulinemia precede the development of type II diabetes in the offspring of diabetic parents. Ann Intern Med 1990;113:909.
85. Martin BC, Warram JH, Krolewski AS, et al. Role of glucose and insulin resistance in development of type 2 diabetes mellitus: results of a 25-year follow-up study. Lancet 1992;340:925.
86. Charles MA, Fontbonne A, Thibult N, et al. Risk factors for type 2 diabetes in white population: Paris prospective study. Diabetes 1991; 40:796.
87. Sicree RA, Zimmet PZ, King HOM, et al. Plasma insulin response among Nauruans: prediction of deterioration in glucose tolerance over 6 yr. Diabetes 1987;36:179.
88. Kadowaki T, Miyake Y, Hagura R, et al. Risk factors for worsening to diabetes in subjects with impaired glucose tolerance. Diabetologia 1984;26:44.
89. Saad MF, Knowler WC, Pettitt DJ, et al. A two-step model for development of non–insulin-dependent diabetes mellitus. Am J Med 1991;90:229.
90. Weyer C, Bogardus C, Mott DM, et al. The natural history of insulin secretory dysfunction and insulin resistance in the pathogenesis of type 2 diabetes mellitus. J Clin Invest 1999;104:787–794.
91. Taylor SI, Accili D, Imai Y. Perspectives in diabetes: insulin resistance or insulin deficiency: which is the primary cause of type 2 diabetes? Diabetes 1994;43:735.
92. Lillioja S, Nyomba BL, Saad MF, et al. Exaggerated early insulin release and insulin resistance in a diabetes-prone population: a metabolic comparison of Pima Indians and Caucasians. J Clin Endocrinol Metab 1991;73:866.
93. Leahy JL, Cooper HE, Deal DA, et al. Chronic hyperglycemia is associated with impaired glucose influence on insulin secretion: a study in normal rats using chronic in vivo glucose infusions. J Clin Invest 1986;77:908.
94. Unger RH. Lipotoxicity in the pathogenesis of obesity-dependent NIDDM: genetic and clinical implications. Diabetes 1995;44:863.
95. Boden G. Fatty acids and insulin resistance. Diabetes Care 1996; 19:394.
96. Boden G, Chen X. Effects of fatty acids and ketone bodies on basal insulin secretion in type 2 diabetes. Diabetes 1999;48:577.
97. Paolisso G, Gambardella A, Amato L, et al. Opposite effects of short- and long-term fatty acid infusion on insulin secretion in healthy subjects. Diabetologia 1995;38:1296.
98. Pickup JC, Mattock MB, Chusney GD, et al. NIDDM as a disease of the innate immune system: association of acute-phase reactants and interleukin-6 with metabolic syndrome X. Diabetologia 1997; 40:1286–1292.
99. Pickup JC, Crook MA. Is type II diabetes mellitus a disease of the innate immune system? Diabetologia 1998;41:1241–1248.
100. Schmidt MI, Duncan BB, Sharrett AR, et al. Markers of inflammation and prediction of diabetes mellitus in adults (Atherosclerosis Risk in Communities Study): a cohort study. Lancet 1999;353: 1649–1652.
101. Fernandez-Real JM, Ricart W. Insulin resistance and inflammation in an evolutionary perspective: the contribution of cytokine genotype/phenotype to thriftiness. Diabetologia 1999;42:1367–1374.
102. Vozarova B, Weyer C, Lindsay RS, et al. High white blood cell count is associated with a worsening of insulin sensitivity and predicts the development of type 2 diabetes. Diabetes 2002;51:455–461.
103. Weyer C, Yudkin JS, Stehouwer CDA, et al. Humoral markers of inflammation and endothelial dysfunction in relation to adiposity and in vivo insulin action in Pima Indians. Atherosclerosis 2002;161: 233–242.
←↑→