Insulin resistance and hypertension: lessons learned from studies in children

In recent years, the concept that insulin resistance and hyperinsulinemia may contribute to pathogenesis of hypertension and cardiovascular morbidity has gained enormous interest. The supporting evidence for the role of insulin resistance comes from several epidemiological and clinical studies. First, hypertensive subjects have been shown to be hyperinsulinemic and insulin resistant compared to normotensive subjects [1,2]. Second, a number of studies have shown a positive correlation between blood pressure and plasma insulin in hypertensive as well as normotensive subjects [3,4]. Third, blood pressure changes during exercise [5] and microvascular function [6] are strongly correlated with insulin sensitivity, indicating that resistance to insulin may influence haemodynamics through effects on vascular reactivity. Fourth, insulin is an independent determinant of left ventricular mass and geometry in untreated patients who have essential hypertension and normal glucose tolerance [7]. Fifth, hyperinsulinemia may predict future blood pressure [8] and cardiovascular risk [9] in healthy subjects independent of other risk factors. Finally, short-term intervention studies [10–12] have identified multiple effects of insulin on the kidneys, autonomic system and vasculature which could lead to hypertension if these effects were sustained. Taken together, these findings suggest a causal relationship between hyperinsulinemia, hypertension and cardiovascular disease, and also that insulin may be a physiological determinant of blood pressure. In contrast to such supporting evidence, there are also studies questioning the importance of insulin resistance per se in causing hypertension. A number of epidemiological studies failed to confirm an independent link between plasma insulin and blood pressure [11,13]. Furthermore, many subjects are normotensive despite being insulin resistant and/or hyperinsulinemic [13,14]. Long-term intervention studies, in contrast to those of short duration, do not support the insulin concept of hypertension [11,15]. Thus, the role of insulin resistance and hyperinsulinemia in the pathogenesis of hypertension remains controversial. Obesity, typically, has been viewed as a major confounder in the relationship between insulin and blood pressure [11,15]. The ‘metabolic syndrome’ is defined as a cluster of cardiovascular risk factors including glucose intolerance, hypertension, low high-density lipoprotein (HDL) cholesterol and hypertriglyceridemia. Most evidence for inclusion of these features in the syndrome is derived from cross-sectional studies. A recent prospective study analysing the contribution of these factors over time indicated that obesity was the central feature of the syndrome [16]. Our understanding of the relative contribution of insulin resistance to blood pressure is also complicated by other factors, including ageing, co-existing diseases and treatment. All these factors may influence insulin resistance. Furthermore, the value of studies in adults, especially those with hypertension, is limited by difficulties in determining whether specific abnormalities in insulin metabolism represent a cause or consequence of the hypertensive state [17]. Therefore, studies in children, before the development of hypertension, might provide important insights into the link between insulin resistance, blood pressure and cardiovascular risk. In this issue of the journal, Sinaiko et al. [18] describe the relationship between blood pressure and insulin resistance in children. Previous studies in children used fasting insulin as an estimate of insulin resistance. An important strength of the present study is the use of an insulin clamp, which is currently the best available standard technique for measuring insulin sensitivity [19]. Blood pressure was not related to insulin sensitivity. The apparent correlation between systolic blood pressure (SBP) and fasting insulin was not significant after adjustment for body mass index. However, the authors demonstrated a significant clustering effect for fasting insulin, body mass index, triglycerides, and HDL cholesterol, which was related to blood pressure grouping (SBP above median versus SBP below median). The authors concluded that the level of blood pressure in children is mediated through body fat, and that insulin resistance, as determined by an insulin clamp, does not play a primary role in blood pressure control at this age of development. On the other hand, the clustering effect of the risk factors according to SBP grouping is consistent with an early relationship of blood pressure to the insulin resistance syndrome. Initially, these conclusions might appear to be contradictory. Why, despite the clustering of the components of metabolic syndrome with predisposition to hypertension, was blood pressure not related to insulin sensitivity? There are several potential factors that should be taken into account. First, the lack of an independent correlation might be due to the method of blood pressure measurement. In the study of Sinaiko et al., blood pressure was defined as the average of two office measurements. We [20,21] and others [22] have shown that the more precise the method of measurement, the closer the relationship between blood pressure and insulin sensitivity. The authors agree that obtaining additional blood pressure measurements would reduce variability and possibly improve the precision of the analyses. Second, the blood pressure responses to hyperinsulinemia represent a complex interplay between the sympathetic nervous system, vessel function/structure and several biochemical mediators, including nitric oxide and leptin [10,13,15,23,24]. Insulin has been shown to evoke both pressor (sympathetic–excitatory) and depressor (vasodilator) actions. With physiological insulin levels in normal humans, the sympathetic–excitatory and vasodilator actions are balanced and blood pressure does not increase [13]. Therefore, the buffering vasodilator mechanisms may overcome any pressor effect of hyperinsulinemia in children. It may be that future structural vascular changes, endothelial dysfunction, coexisting diseases or genetic predisposition to hypertension, would augment the sympathetic (pressor) action of insulin or attenuate the vasodilator (depressor) action, thereby permitting insulin to produce a sympathetically mediated increase in arterial pressure. Hausberg et al. [25] provide evidence for an age-related differential response to insulin. They studied the effects of insulin in young versus elderly healthy subjects. While insulin caused the expected vasodilation in young subjects, elderly subjects responded to insulin by vasoconstriction. This vasoconstriction was not explained by greater sympathetic activation, since the increase in sympathetic activity in the elderly subjects was less than that seen in the young subjects. The natural history of the insulin resistance syndrome might depend on several genetic and environmental factors. Support for the importance of genetic factors comes from the observations that propensity to obesity and hypertension differ substantially among different populations [13]. Furthermore, the roles of hyperinsulinemia and increased sympathetic nervous system activity as mediators for the relationship between obesity and hypertension differ between various ethnic groups [26]. Mark and Anderson [13] introduced the concept of blood pressure ‘sensitivity or resistance’ to insulin resistance and hyperinsulinemia. They suggested that genetic factors may be decisively influential in this effect. According to this concept, insulin resistance under normal conditions is not sufficient to influence blood pressure. However, under certain circumstances, hyperinsulinemia may influence blood pressure in vulnerable subjects. In particular, age, increased body weight and prolonged exposure to hyperinsulinemia may contribute to future hypertension. Traditionally, insulin resistance has been linked to volume expansion, sodium retention and enhanced sympathetic nervous system activity. Recent evidence suggests that insulin resistance, especially if associated with obesity, may increase cardiovascular risk through increased activity of the systemic renin–angiotensin system [27] and subclinical inflammation, as estimated by C-reactive protein [28]. Indeed, it has been suggested that a direct link between insulin resistance and blood pressure control may not be the best way to explain the role of insulin resistance in cardiovascular morbidity and mortality [15]. The early link between insulin resistance and other risk factors in children predisposed to hypertension may grow stronger as these children transition from adolescence into adulthood. A recent study [29] in apparently healthy, non-obese individuals has shown that age-related clinical events, defined as hypertension, coronary heart disease, stroke, cancer or type 2 diabetes, developed in 36% of individuals in the upper tertile of insulin resistance at baseline, who were followed for an average of 6 years, whereas no clinical events were observed in the most insulin-sensitive tertile. There is increasing evidence that hyperinsulinemia might be also linked to congestive heart failure [30] and obstructive sleep apnoea [31]. It is clear that insulin resistance is a complex abnormality and no single mechanism can be identified as the dominant aetiologic factor. The study by Sinaiko et al. further confirms that the relationship between insulin resistance and blood pressure is not straightforward. The clustering of features of the metabolic syndrome in subjects predisposed to hypertension might have important implications for prevention, particularly with regard to whether interventions targeted at insulin resistance would have beneficial effects on the other components of the syndrome. This may be especially relevant when the intervention is introduced at an early age.