On the paradox insulin resistance/insulin hypersensitivity and obesity: two tales of the same history

Abstract

Insulin resistance (IR) associated with obesity represents a well-known risk factor for chronic disease. IR development may occur to hinder stressful conditions to provide an appropriate energetic supply to non-insulin-sensitive tissues. However, conditions of stress turn out to be ‘maladaptive’ in the long term, leading to chronic diseases. Paradoxically, insulin hypersensitivity and/or hypersecretion causing post-prandial hypoglycemia resulting in increased food intake and weight gain, can represent an event preceding obesity and IR. By performing an OGTT in obese or obese-prone individuals we observed that tardive post-prandial hypoglycemia (3h from glucose load) is not a rare event (32%); in 12% of cases it paralleled with low insulin levels, resulting in the ‘true insulin hypersensitivity’. By using Matsuda-method, we confirmed the presence of insulin hypersensitivity in this group. Therefore the early recognition of this phenomenon could be useful as a predictive biomarker to identify patients prone to develop obesity and obesity related-disorders.

Keywords::

• cardiovascular risk
• insulin hypersensitivity
• insulin paradox
• insulin resistance
• obesity

Alice: “This is impossible...”

The Mad Hatter: “Only if you believe it is.”

–“Alice in Wonderland”

Obesity incidence is increasing worldwide, reaching an epidemic rate. Obesity is a well-recognized risk factor for the development of cardiovascular diseases, such as coronary artery disease and heart failure, and also for Type 2 diabetes mellitus and cancer, driving the obesity-attributable death rate to around 3 million per year [1].

The development of insulin resistance (IR) associated with obesity, together with an inflammatory state, has been considered the pivotal phenomenon that can trigger a series of physiopathological events that may culminate in increased cardiovascular morbidity and mortality [2]. Physiologically, after a meal, insulin is released from pancreatic β cells, which constitutes the main anabolic hormone that regulates substrate utilization and storage in multiple tissues, including the liver, skeletal muscle, adipose tissue and the heart [3]. During fasting, when insulin levels are low, the substrates stored become available to maintain normal plasma glucose concentrations, for example, to prevent hypoglycemic events.

IR: systemic, hepatic & cardiac

When inefficient glucose uptake by muscle and adipose tissue occurs, a compensatory insulin hyperproduction is necessary to compensate for high plasma glucose concentrations. Moreover, during chronic hyperinsulinemia, secondary to IR to the glucose pathway, the relative normal insulin sensitivity to the lipid pathway could lead to increased fat storage at the peripheral level (adipose tissue and skeletal muscle) and in the liver.

Hepatic IR appears to be a selective process [4]: in this condition, insulin fails to suppress gluconeogenesis but continues to promote lipogenesis of free fatty acids and their esterification in triglycerides, leading to hepatic increased lipid storage (hepatic steatosis). This condition is associated to an excessive release of triglycerides into the bloodstream. Thus, the detrimental combination at the systemic level of hyperglycemia and hypertriglyceridemia heralds the state of IR.

The heart is not immune to the deleterious effects of IR. A central point in the evaluation of myocardial IR is to be able to distinguish between the primary metabolic effects of peripheral IR (e.g., hyperinsulinemia, hyperglycemia and hyperlipidemia) from the secondary changes of insulin signaling, independent of the humoral milieu, that could be intrinsically altered at the myocardial level, for example, the true myocardial IR (Figure 1). It is not easy to discern in vivo between the primary and the secondary myocardial IR. IR-associated dyslipidemia promotes elevated rates of myocardial fatty acid uptake and oxidation, while glucose oxidation rate is depressed with the consequent decrease in cardiac efficiency [5]. Elevated levels of fatty acids may cause an excessive accumulation of lipid intermediates, such as diacylglycerols and ceramide, which could induce myocardial IR and myocyte apoptosis [6].

Figure 1. Mechanisms and consequences of insulin resistance and the cardiovascular system.

Other negative effects of IR/hyperinsulinemia on cardiovascular disease include the potential role of insulin on essential hypertension. Noteworthily, high insulin levels can contribute to deterioration of cardiac and renal function by antinatriuretic effect [7] and have stimulating effects on sympathetic nervous system activity [8].

Is IR an adaptive or maladaptive mechanism? can IR operate as a protective factor?

It has been demonstrated that IR is directly associated with obesity, diabetes and cardiovascular disease, entailing an increased risk of mortality. Nevertheless, from an evolutionary point of view, the development of IR may have a protective role during physiological or pathological states associated with negative energy balance, such as starvation, infection or trauma, to assure sufficient glucose provision toward the energy-demanding brain [9]. In this setting, the development of IR that occurs during physiological states, such as pregnancy, could be considered a protective phenomenon able to maintain a proper energetic supply to the fetus during body growth [10]. Interestingly, it has been suggested than even in some pathological conditions, such as myocardial infarction, obesity (and probably, by extension IR) could be advantageous in terms of size of myocardial infarct [11] and mortality [12,13].

IR: from adaptive mechanism to disease state

Which events can transform an adaptive, protective, physiological response to stressful conditions into a pathological element capable of producing a chronic devastating disease?

As shown in Figure 2, it can be hypothesized that the allostatic load (i.e., the ‘point of rupture’) of the system could be reached by repetitive or prolonged exposure to stressful variables, such as multiple and frequent exposure to high-calorie diets succeeding infections or aging. The relative irreversibility of the process is suggested by the enormous difficulty to return to the ‘healthy status,’ observed in several conditions associated with IR, such as obesity, diabetes mellitus, essential hypertension and cardiovascular disease. These key mediators of obesity-induced metabolic disease have been recently defined as ‘evolutionarily conserved adaptive traits with maladaptive effects in the modern obesogenic environment’ [14]. The above concept calls for our maximal attention to improve our diagnostic skills to be used at a point where the process could be yet ‘reversible.’ As observed in Figure 2, a potential target for the prevention of IR obesity would necessarily be a correct identification of early predictors of IR-associated diseases, such as insulin hypersensitivity.

Figure 2. Representation of the hypothetic chain of events that lead from insulin hypersensitivity to insulin resistance-associated diseases. On the basis of genetic predisposition, the early events are initially displayed by insulin hypersensitivity and postprandial hypoglycemia. Postprandial hypoglycemia leads to increased caloric consumption and, therefore, to body weight increase. A new steady state of insulin sensitivity is reached to maintain stable body weight. With iterative exposure to the deleterious stimulus (frequency factor) and/or the duration of exposure is extended (time-exposure factor), an allostatic load is reached and insulin resistance is transformed from an adaptive to a maladaptive mechanism.

Insulin hypersensitivity, a potential early biomarker of obesity development

Needless to say, in any definition that applies a criterion of ‘normality’ or ‘resistance’ to the effect of any hormone, a subgroup of ‘hypersensitive’ individuals to the putative hormonal effects must exist. The physiological adaptations and pathological states that could result from changes in target tissue sensitivity to the action of hormones are of extreme importance in normal and pathological states. Hypersensitivity syndromes due to prereceptor, receptor and postreceptor abnormalities have been described in endocrine physiology. In these settings, the presence of a hypersensitive state to the hormonal effects of thyroid [15], sexual [16] and glucocorticoid hormones [17] has already been reported.

Hypersensitivity to the physiological action of insulin had been well described in the 1940s by the prominent physiologist, the Nobel Prize winner Bernardo Houssay [18]. It is astonishing that compared with IR research, insulin hypersensitivity has received very poor scientific attention. By applying the keywords ‘insulin’ and ‘hypersensitivity’ in Medline Scholar Google and after excluding animal studies, we have found less than 50 published articles in the past 25 years compared with the nearly 763,000 references found during the same period when the keyword ‘insulin resistance’ was used.

By definition, insulin hypersensitivity must be suspected in the presence of combined hypoglycemia associated with normal or low plasma insulin levels. Therefore, the key diagnostic element of insulin hypersensitivity is heralded by the presence of hypoglycemia, usually postprandial. This condition is elicited by the ingestion of carbohydrates and/or can be diagnosed using an oral glucose tolerance test. The usual threshold for the diagnosis of postprandial hypoglycemia is 70 mg/dl (or 3.9 mmol/l) [19]. Hypoglycemia may occur earlier, that is, in the first 2 h after oral glucose load (and it is usually secondary to some well-known disorders such as hyperthyroidism or gastric surgery), and tardive, that is, 3–5 h after glucose load or meal stimulus. In more than 50% of the cases, tardive postprandial hypoglycemia could be secondary to insulin hypersensitivity [19].

Many authorities have denied the pathological role of postprandial hypoglycemia, at least when this alteration arises in the absence of symptoms. This assumption is, however, in sharp contrast with other well-accepted clinical entities that similarly occur in the absence of symptoms, such as hypoglycemia secondary to the use of insulin or antidiabetic drugs, asymptomatic hyperglycemia and hypercalcemia. When symptomatic postprandial hypoglycemia occurs, it presents with manifestations of increased sympathetic activity, such as anxiety, tremor, dizziness, tachycardia, weakness, increased perspiration and hunger. The symptoms subside with carbohydrate ingestion, but it may be oligosymptomatic or completely asymptomatic.

The mechanism behind tardive hypoglycemia has been predominantly attributed to increased first phase of insulin secretion secondary to IR. As defined above, this condition is not ‘true’ insulin hypersensitivity, for example, a hypoglycemia episode that occurs in the presence of normal or low plasma insulin levels.

By evaluating insulin and glucose responses during an oral glucose test tolerance (OGTT) performed in our research hospital in a large group of obese and obesity-prone individuals (lean or overweight subjects with a strong family history of obesity) (n = 941) we were able to identify two different types of tardive postprandial hypoglycemia. All patients were not taking any medications potentially affecting insulin sensitivity (Figure 3B & 3D). The first group showed elevated insulin levels during OGTT (that we have termed ‘hypoglycemia secondary to hyperinsulinemia and IR, Figure 3A & 3B, blue line) and the second group exhibited normal or low levels of plasma insulin (that we have termed ‘true’ insulin hypersensitivity group’, Figure 3C & 3D, blue lines and circles). Overall, nearly more than a quarter (31%) of the whole population showed tardive hypoglycemia, while 10% of the entire examined population had true insulin hypersensitivity. Interestingly, we observed this late condition predominantly in lean patients prone to the development of obesity.

Figure 3. Graphical results of 941 oral glucose test tolerance (OGTT) studies, according to glucose and insulin response to oral glucose load. Insulin-resistant OGTTs: (A) Insulin curves of normal and insulin-resistant subjects. Blue line indicates insulinemic response of subjects with hypoglycemia (<3.9 mmol/l). (B) Glucose curves of subjects with high insulin responses during OGTT. Red lines indicate normal glucose responses. Insulin-sensitive OGTTs. (C) Insulin curves of subjects with normal and insulin hypersensitivity. Blue lines and circles indicate insulinemic responses of patients with hypoglycemia and low plasma insulin levels (<70 mU/ml at 1 h after glucose load). (D) Glucose curves of subjects with low insulin responses during OGTT. Red lines indicate normal glucose responses.

Notably, the insulin sensitivity index, as assessed using the Matsuda method [20], was threefolds higher in low-insulin hypoglycemic patients than the cut-off for IR and it allows to discriminate between IR and insulin hypersensitivity in patients with postprandial hypoglycemia (Figure 4).

Figure 4. Insulin sensitivity index by the Matsuda method: ‘true’ insulin hypersensitive group (yellow bar); hypoglycemic–hyperinsulinemic group (blue bar); and hyperglycemic groups (red bars).

We believe that postprandial hypoglycemia is a ‘true’, underestimated and in many cases, neglected disorder, that can lead to increased chronic caloric consumption and then to obesity. In this regard, it is intriguing to note that in some classic longitudinal studies analyzing risk factors for the development of obesity, increased insulin sensitivity (as measured by the glucose clamp technique) emerges as a potent predictor of obesity [21]. Accordingly, in the EGIR study, a large number of obese individuals exhibited high or normal insulin sensitivity [22]. Also, in a recent article, it has been reported that obesity-prone women were more insulin sensitive compared with a group of lean, non-obesity-prone subjects [23]. In line with these findings, increased insulin sensitivity was found in a group of normal-weight, healthy men with a strong familial history of obesity, [24], suggesting that increased insulin sensitivity could be genetically determined and could be an early predictor of weight gain.

As observed in Figure 3, insulin hypersensitivity could be seen predominantly after 2 h of oral glucose stimulus. This finding implies that it could be useful to extend the OGTT study to at least 3 h to diagnose a potential hypoglycemic event. We also suggest that insulin level determinations must be added to the usual, simplistic approach of determining plasma glucose concentrations only.

In conclusion, our findings obtained from a large dataset of OGTTs show that postprandial hypoglycemia is not a rare event and could be found in both IR and insulin sensitivity states. Postprandial hypoglycemia could be clinically significant in terms of prediction value for the development of obesity. Early recognition of postprandial hypoglycemia could represent a useful biomarker to identify patients with predisposition to obesity and obesity-related disorders that can culminate in increased cardiovascular mortality.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.

Notes