1. Introduction

Subjects with insulin resistance and hyperinsulinemia are at high risk of developing obesity, type 2 diabetes, cardiovascular disease, cancer and premature mortality ^{[1][2][3][4][5][6][7][8][9]}[1,2,3,4,5,6,7,8,9]. The dogma has been for many years that insulin resistance is first and precedes hyperinsulinemia and type 2 diabetes mellitus. In this view, hyperinsulinemia was secondary and represented a compensatory mechanism to overcome systemic (peripheral) insulin resistance [10]. Insulin resistance was considered to be the primary etiological factor in the development of obesity, type 2 diabetes, cardiovascular disease and cancer, whereas the compensatory hyperinsulinemia was thought to be a direct consequence of insulin resistance [11]. However, recently the correctness of this view has been called into question ^[12][13][14][12,13,14]. It has even been proposed that hyperinsulinemia per se is primary and causes (and is not a consequence of) insulin resistance [13]. In this new concept, insulin resistance is proposed to be a physiological defense mechanism of the body that tries to prevent the development of hypoglycemia and to protect critical tissues from metabolic stress and nutrient-induced injury ^[15][16][15,16]. This opens the possibility that (early) interventions able to normalize/reduce plasma insulin concentrations might play a key role in the prevention and treatment of obesity, type 2 diabetes, cardiovascular disease, cancer and premature mortality ^[10][17][10,17].

3. Loss of Pulsatile Insulin Secretion Contributes to Insulin Resistance

In healthy subjects, insulin secretion into the circulation is pulsatile and normally accounts for 75% of the daily insulin secretion ^[24][58]. It has been proposed that pulsatile insulin release may avoid the downregulation of (hepatocyte) insulin receptors compared to the constant delivery of insulin ^[25][63]. Prolonged near-physiological pulsatile insulin infusion has a greater hypoglycemic effect than continuous insulin infusion ^[26][64]. Moderate pulsatile hyperinsulinemia in nondiabetic human subjects does not induce insulin insensitivity ^[26][64]. This is in contrast to what is found after continuous hyperinsulinemia ^[26][64]. It has been demonstrated that hyperinsulinemia produced by continuous infusion of insulin (which increased plasma insulin concentrations to levels similar to those observed in insulin-resistant conditions) can produce insulin resistance (decreased insulin receptor sensitivity), and this decrease in insulin action may occur at the receptor and post-receptor level of the insulin receptor and is tissue-specific ^[27][65]. The loss of pulsatile insulin secretion is an early feature in the development of type 2 diabetes and may be involved in the (patho)genesis of insulin resistance in a variety of circumstances ^[27][28][65,66]. Intravenous delivery of insulin in a constant versus pulsatile pattern led to delayed activation of hepatic insulin receptor substrate (IRS)-1 and IRS-2 signaling, impaired activation of downstream insulin signaling effector molecules AKT and Foxo1, and decreased expression of glucokinase, suggesting that the physiological pulsatile pattern of insulin delivery is important for normal hepatic insulin signaling and glycemic control and essential to preserve insulin sensitivity ^[29][67]. In addition, several lines of evidence suggest that the pattern of insulin secretion by the pancreas determines hepatic insulin clearance: the liver preferentially extracts insulin delivered in pulses ^[30][68]. Therefore, the pulse mass of insulin release dictates both hepatic (directly) as well as extrahepatic (peripheral) insulin delivery ^[30][68]. Animal studies have also demonstrated the importance of pulsatile insulin delivery in the development of insulin resistance [16]. Although an acute rise in insulin is stimulatory for the insulin receptor, persistently and continuously elevated insulin levels desensitize the insulin receptor through multiple mechanisms, both at receptor and post-receptor level ^{[12][31][32][33]}[12,69,70,71]. Continuous and long-term exposure to insulin causes a reduction in the number of insulin receptors at the cellular surface by promoting internalization as well as degradation of insulin receptors ^[31][69]. With continuous and long-term exposure to insulin, the kinase activity of the insulin receptor diminishes, probably because of combined effects of phosphorylation on serine residues on the insulin receptor, the dephosphorylation of tyrosines by the action of phosphatases, and the binding of inhibitory molecules ^{[12][34][35][36][37]}[12,72,73,74,75]. All of these effects downregulate insulin receptor signaling and thereby cause insulin resistance.

4. Hyperinsulinemia Precedes Insulin Resistance

In healthy subjects, plasma glucose is maintained within narrow ranges by a classic negative feedback system ^[38][76]. After meals, insulin secretion by the pancreas is stimulated by a rise in glucose and this brings plasma glucose back to baseline ^[38][76]. In this negative feedback system, insulin secretion is controlled and glucose levels remain within the normal range as long as subjects are able to overcome insulin resistance by increasing insulin secretion ^[38][76]. However, as soon as subjects have lost this ability, they will progress to impaired glucose tolerance and/or type 2 diabetes ^[38][76]. Until very recently, the prevailing view was that insulin resistance (that is, resistance to insulin’s role in promoting glucose uptake by muscle and fat cells) preceded and caused hyperinsulinemia ^[39][77]. In this view, insulin resistance was the initial defect leading to the development of metabolic syndrome, hyperglycemia and type 2 diabetes after years or even decades later ^[40][78]. However, in genome-wide association studies (GWAS), only a few loci point to insulin resistance as the primary cause of type 2 diabetes, while the majority of the loci identified by GWAS point towards defects of the β-cell of the pancreas ^[41][42][55,79]. This raises the distinct possibility that a β-cell defect in insulin secretion that initially causes inappropriate hypersecretion of insulin at basal plasma glucose concentrations may be a driver of insulin resistance by insulin-induced downregulation of insulin receptors ^[13][16][13,16]. It is important to realize that the actual diagnostic criteria for normal glucose tolerance, impaired glucose tolerance and type 2 diabetes are not defined on the basis of pathophysiological abnormalities ^[43][80]. Thus, although cross-sectional studies have postulated that insulin secretion follows an inverted U-pattern (also termed Starling’s curve of the pancreas) during natural progression from normal glucose tolerance to impaired glucose tolerance and type 2 diabetes ^[34][72], it has been repeatedly reported that a large part of individuals with normal glucose tolerance already show hyperinsulinemia before the development of impaired glucose tolerance/obesity. For example, Ferrannini et al. found increased plasma insulin levels in subjects with normal or near normal glucose tolerance ^[43][80]. In addition, it is difficult to understand how insulin resistance in subjects with normal glucose tolerance could be responsible for increased insulin secretion when blood glucose concentrations are still within the normal range. This is another argument against a primary role for insulin resistance-mediated hyperinsulinemia ^[44][81]. In addition, if fat cells were already resistant to insulin in the early phase, it is difficult to understand how hyperinsulinemia could stimulate lipogenesis and induce obesity by driving calories into fat cells ^[45][82]. Recently, a new model has been brought forward: in this new model, hyperinsulinemia is considered the primary event that secondarily causes insulin resistance and type 2 diabetes ^[12][44][12,81]. In support of this new model, insulin secretion has been found to be elevated before the development of hyperglycemia in a longitudinal study of Rhesus monkeys, developing a form of type 2 diabetes, which appears to be very similar to that found in humans ^[46][83]. In addition, an increasing number of human studies support the hypothesis that basal hyperinsulinemia is primary and that it contributes secondarily to insulin resistance and many diseases and conditions [12].

In another study, half of the 4485 subjects showed hyperinsulinemia despite normal glucose clearance, suggesting that hyperinsulinemia likely occurs as a “silent disease” in a substantial proportion of an otherwise healthy population ^[47][85]. In Pima Indians with normal glucose tolerance, evidence was found that fasting hyperinsulinemia itself, independent of a low rate of insulin-stimulated glucose uptake, predicted the cumulative incidence of type 2 diabetes during a 7 year follow-up ^[48][86]. Therefore, the authors concluded that high fasting plasma insulin concentrations were not a reflection of insulin resistance but rather the consequence of a basal hypersecretion of insulin relative to the degree of insulin resistance ^[48][86]. In a fourth study, the prevalence of insulin hypersecretion in nondiabetic, normotensive obese women exceeded the prevalence of insulin resistance, which was relatively low ^[49][87].

5. How Can Hyperinsulinemia Be Modified?