Insulin Resistance and Hyperinsulinemia: the Egg and the Chicken

Insulin resistance and hyperinsulinemia are closely associated (1). It is widely recognized that a dynamic relationship exists between insulin secretion and insulin resistance in which worsening insulin action (ie, insulin resistance) triggers an increase in insulin secretion, and the hyperinsulinemia compensates for the defect in insulin action. Based on this relationship between insulin secretion and insulin resistance, the prevailing paradigm of the natural history of type 2 diabetes mellitus (T2DM) was formulated that insulin resistance is the primary defect in individuals destined to develop T2DM, and compensatory hyperinsulinemia ensues to offset the diminished efficacy in insulin action. Hyperglycemia develops only when the β-cell fails to augment its insulin secretion and compensate for insulin resistance (1). Indeed, severe insulin resistance preceding or accompanied by hyperinsulinemia can be detected long before the development of T2DM (2). However, the coexistence of insulin resistance and hyperinsulinemia does not establish a causal relationship between the 2, nor does it define which defect, insulin resistance or increased insulin, precedes the other. Using data obtained in Pima Indians with the euglycemic insulin clamp and oral glucose tolerance test (OGTT), we pointed out in 1997 (1) that the results were consistent with β-cell hypersecretion of insulin as the primary defect followed by the development of insulin resistance secondary to the resultant hyperinsulinemia (3).

In the past several years, other studies (reviewed in [4]) have challenged this “central dogma” of the natural history of T2DM, and suggested that hyperinsulinemia is the primary defect that precedes insulin resistance and triggers the events responsible for the development of T2DM. This discussion about the “egg and chicken” relationship between hyperinsulinemia and insulin resistance has important clinical implications. If hyperinsulinemia is a compensatory response to insulin resistance, then further elevation of plasma insulin concentration becomes a good strategy to overcome insulin resistance to achieve optimal glycemic control, although the hyperinsulinemia may result in other undesired metabolic effects. Conversely, if insulin hypersecretion and hyperinsulinemia is the primary defect that triggers insulin resistance, therapeutic efforts should focus on strategies to reduce plasma insulin concentration to achieve optimal metabolic control in T2DM patients.

In the present issue of the journal, Armiyaw and colleagues (5) provide evidence that challenges the prevailing view of the natural history of T2DM, in which insulin resistance precedes and is the causative factor responsible for the increased insulin secretion. They compared insulin secretion and insulin sensitivity in 2 groups of middle-age patients of African American and non-Hispanic white ethnicity. Insulin sensitivity was measured with the euglycemic hyperinsulinemic clamp and insulin secretion was measured with intravenous glucose tolerance test (IVGTT). The 2 groups had comparable body structure as measured with body mass index, total body fat, and lean mass. The plasma glucose profile during IVGTT was comparable among the 2 groups, yet, insulin secretion was higher by 63% in African Americans compared to whites. Using the insulin clamp with pharmacologic insulin infusion rate (120 mU/m²/min). The rate of insulin-stimulated glucose disposal was found to be similar in African American and non-Hispanic whites. Hepatic insulin sensitivity measured as the product of basal hepatic glucose production rate and the fasting plasma insulin concentration was similar in the 2 groups, as were the fasting plasma insulin and free fatty acid (FFA) concentrations. From these results the authors concluded that African Americans have an augmented insulin secretory response to glucose, that is, insulin hypersecretion, compared to whites, and this insulin hypersecretion was due to enhanced β-cell sensitivity to plasma glucose, that is, an endogenous β-cell phenotype. An OGTT was not performed, so neither glucose nor the plasma incretin hormones could be evaluated.

Although the study of Armiyaw and colleagues was relatively small (n = 36) and cross-sectional in design, which precluded establishing a causal relationship, it has several strengths (1): Insulin sensitivity was measured with euglycemic insulin clamp (2), and insulin secretion was measured with IVGTT on a different day. Thus, independent measurements of insulin sensitivity and insulin secretion were obtained (3). Plasma C-peptide was used to measure insulin secretion during the IVGTT. Thus, differences in insulin clearance could not affect the results. However, some limitations may raise concerns about the interpretation of the results. First, steady-state plasma insulin concentration during the insulin clamp was not measured in the study, and it is likely that it was high very high (> 150 µU/mL) because of the very high insulin infusion rate used. Such high plasma insulin concentration may have obscured insulin resistance that was present under more physiologic plasma insulin concentration. Second, insulin sensitivity in liver and adipocytes was measured during the fasting state and the suppression of hepatic glucose production and lipolysis by insulin were not measured. Because insulin secretion should be interpreted in the context of the prevailing level of insulin resistance, that is, the so called disposition index, a different result will be obtained should the disposition index be calculated on the basis of hepatic insulin resistance or peripheral (muscle) insulin resistance or adipocyte insulin resistance. For example, following short-term (5 days) feeding in healthy nonglucose tolerant individuals, 2 different disposition indices were calculated: one based on peripheral insulin sensitivity and one based on hepatic insulin sensitivity. Because fat overfeeding induced hepatic, but not peripheral, insulin resistance, the disposition index calculated from peripheral (muscle) insulin sensitivity was disproportionally and significantly increased whereas the disposition index calculated from hepatic insulin resistance was not increased (6). Thus, depending on which index of insulin resistance is used, one might or might not conclude that a primary defect of β-cell hypersecretion of insulin is present. Last, if insulin hypersecretion is the primary defect in African American patients, why were they not insulin resistant in the muscle and/or liver? It is noteworthy that despite a marked increase in glucose-stimulated insulin secretion during the IVGTT in African Americans, the fasting plasma insulin concentration was comparable among the 2 groups. This might suggest that the insulin hypersecretion is still in early stages and could provide a possible explanation for the lack of insulin resistance in skeletal muscle.

Let us assume that insulin hypersecretion is the primary defect that initiates the sequence of events responsible for T2DM. What then, is the mechanism responsible for that? A primary defect in β cells causing hyperresponsiveness to nutrient stimuli can result in chronically elevated plasma insulin concentration in response to increased food intake. Increased food intake is widely recognized as a major factor driving the current epidemic of obesity and diabetes. Consistent with this hypothesis, fasting and postprandial hyperinsulinemia have been shown to precede the development of insulin resistance in individuals with juvenile-onset obesity, and in a variety of animal models of obesity and diabetes including ventromedial hypothalamus lesions in normal rats and the fa/fa rat (7). According to this hypothesis, alterations in central nervous system function, for example, neuropeptide Y, or autonomic nervous system function, result in increased food intake and increased insulin secretion. Previous studies have demonstrated that a small chronic elevation in plasma insulin concentration in lean healthy individuals results in insulin resistance (3). Thus, a primary defect in β cells causing hyperresponsiveness to nutrient stimuli and/or increased insulin secretion in response to altered central nervous system function can result in chronically elevated plasma insulin concentration and trigger insulin resistance resulting in insulin resistance and hyperinsulinemia.

It should be noted that cross-sectional studies reporting differences in insulin secretion for the same level of insulin resistance do not necessarily prove or disprove a dissociation in the relationship between insulin sensitivity and insulin secretion or which defect came first, insulin resistance or hyperinsulinemia. The inverse relationship between insulin secretion and insulin sensitivity is curvilinear. As anticipated from all biological phenomena, a certain degree of variability will exist among the function describing the quantitative relationship between the 2. Thus, multiple curves will describe the relationship between insulin secretion and insulin sensitivity among different populations. Thus, differences in insulin secretion among the different groups for the same level of insulin sensitivity measured at one point of time, as reported by Armiyaw and colleagues (5) in the present study, do not necessarily reflect a dissociation in the relationship between insulin secretion and insulin sensitivity, rather it could indicate that the curve that governs the dynamic relationship between insulin secretion and insulin sensitivity in each ethnic group operates at a different percentile of the total population.

A major limitation of the hypersecretion hypothesis is the lack of convincing evidence linking early insulin hypersecretion to increased risk of future T2DM. Longitudinal studies are required to establish such a causal relationship. Longitudinal data from Pima Indians (8) that examined insulin sensitivity with the euglycemic insulin clamp and insulin secretion with OGTT and IVGTT have demonstrated that individuals who progressed to T2DM had similar insulin sensitivity to those who did not progress to T2DM. Acute insulin secretion measured with IVGTT was comparable among the 2 groups. However, fasting and 2-hour plasma insulin concentration during the OGTT were significantly higher in progressors than in nonprogressors. Over time, insulin resistance worsened and insulin secretion decreased in progressors, whereas insulin secretion and insulin sensitivity both were maintained in individuals who remained nonglucose tolerant. Although these findings could imply that high-risk individuals (progressors) manifested hyperinsulinemia for a comparable level of insulin sensitivity compared to low-risk individuals, it is not possible to determine the cause of hyperinsulinemia in that study because insulin secretion was measured with the plasma insulin concentration. Thus, differences in insulin clearance could confound the results. Further, the discrepancy in plasma insulin concentration between IVGTT and OGTT suggests that differences in incretin hormones could account, at least in part, for some of the differences. Thus, more longitudinal studies are needed to establish the relationship between insulin hypersecretion and risk of T2DM.

Interventional studies could provide a strategy to establish the causal relationship between insulin resistance and hyperinsulinemia. Inducing insulin resistance in healthy individuals is one possible strategy. Interventional studies that produce insulin resistance (eg, elevation of plasma FFA concentrations) in lean healthy individuals is accompanied by hyperinsulinemia. However, it could be argued that elevated plasma FFA levels affect insulin sensitivity and insulin secretion directly. Examining the effect of inhibition of insulin secretion in patients with insulin hypersecretion, for example, with diazoxide treatment, on insulin sensitivity could be another strategy to establish the causal relationship between insulin resistance and hyperinsulinemia.

In summary, although the cause/effect relationship between insulin resistance and hyperinsulinemia is still debated, there is a consensus that this combination is detrimental to health and associated with worse health outcomes. It is a precursor of T2DM and is associated with other metabolic abnormalities, that is, metabolic syndrome, all of which are risk factors for cardiovascular disease. Thus, interventions that reverse insulin resistance and hyperinsulinemia should be introduced in individuals who manifest this combination.

Additional Information

Disclosure Summary: The authors have nothing to disclose.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.