Insulin resistance: metabolic and somatic changes in children

. Background. Insulin resistance is the major sign of etiology and pathogenesis of type 2 diabetes mellitus and metabolic syndrome and can precede its development for many years. Early identifying the beginning of insulin resistance in children is important to prevent diabetes mellitus in adult life. The purpose was to identify metabolic and somatic changes in children with insulin resistance. Material and methods. Out of 182 children of the general sample, who was estimated fasting plasma insulin and glucose, HOMA-IR, and glucose/insulin ratio, 2 groups were formed: group 1 — children with IR — 56 (30.8 %) and group 2 — 126 (69.2 %) children with normal insulin sensitivity. In children anthropometric data, lipid metabolism (total cholesterol, triglycerides, HDL-C, LDL-C, VLDL-C), blood pressure, leptin were determined. Results. From examined subjects 56 children were generally obese (BMI > 95 th percentile), 71 children were abdominally obese (WC > 90 th percentile), 55 children were with normal body mass (BMI < 90 th percentile). Insulin resistance was identified in 21 (37.5 %) children with general obesity more rarely, than in 38 (39.4 %) children with abdominal obesity (p = .049) and in 7 (12.7 %) children with normal BMI (p = .003). In insulin-resistant children BMI, waist and hip circumference was larger than in children with normal insulin sensitivity. The lipid profile in children with different insulin sensitivity did not differ, but in insulin-resistant children an association of basal glucose with TG/HDL-C ratio (r = .53; p = .001), blood insulin with TG (r = .34; p = .018), and TG/HDL-C ratio (r = .54; p = .001) was estimated. The HOMA-IR significantly correlated with VLD-C (r = .40; p = .005), TG (r = .49; p = .001), TG/HDL-C ratio (r = .43; p = .002). The glucose/insulin ratio was in significant association with the TG/non-HDL-C ratio. The incidence of hypetension (> 95 th percentile) diagnosis in insulin-resistant children was by 33.8 % higher (p = .001). Blood leptin concentration was 1.8 falled higher in insulin-resistant children and significantly correlates with waist circumference, fasting insulin, HOMA-IR, and diastolic blood pressure. Conclusions. Insulin resistance is related to cardiometabolic risks, such as general and abdominal obesity, hypertension, dyslipidemia, hyperleptinemia, and leptin resistance, and is a screening biomarker for children and adolescents with an increased risk of cardiometabolic diseases.


Introduction
The widespread prevalence of type 2 diabetes mellitus (DM) and metabolic syndrome (MS) in the adult population and understanding that adult illnesses begin in childhood requires the early diagnosis of functional abnormalities of carbohydrate metabolism in children and their correction. Insulin resistance (IR) is one of the major signs of etiology and pathogenesis of type 2 DM and MS and can precede its development for many years [1][2][3][4][5]. Identifying the beginning of IR in children is important to prevent DM in the future [6].
The novel data indicate that IR and DM are common among children and adolescents [7] and increase in the children population [8]. In the meta-analysis, it was found that children's prevalence rates of IR are between 3.1 in Greece [9] and 44 % in Pacific Island teenagers (New Zeeland) [10], with the gender higher difference in obese boys -68.4 % [10]. However, it was also determined that a higher prevalence of IR was in girls than in boys, partly due to the difference in pubertal development [11].
It was estimated that changed carbohydrate metabolism and IR are estimated in 35 % children with abdominal obe-Mìžnarodnij endokrinologìčnij žurnal, ISSN 2224-0721 (print), ISSN 2307-1427 (online) Оригінальні дослідження / Original Researches sity and one MS criterion, in 60.7 % children with abdominal obesity and two MS criteria and in 75 % children with abdominal obesity and three MS criteria [12].
IR was formed in the process of evolution of human beings as a way to adapt to periods of prolonged food deficit and is a physiological norm in certain periods of life such as puberty. It was demonstrated that IR occurs both as a result of primary irreversible violations and as a result of secondary reversible factors. IR etiology can be divided into acquired, hereditary, and mixed. The great majority of people with IR fall into the acquired categories [13].
IR occurs imperceptibly gradually at an early age. In the beginning, IR is asymptomatic, and metabolic and functional changes are identified only during laboratory and instrumental examinations. Type 2 DM and MS develop in the late stages of IR when it is sometimes difficult to get back the disease [13].
Insulin is an anabolic hormone that plays an impor tant role in the regulation of glucose, lipid homeostasis, and ener gy storage through its natural metabolic effects on classic insulin-responsive tissues: adipose tissue, liver, and skeletal muscle. Insulin increases glucose transport across the cell membrane, by the rise of hexokinase and 6-phosphofructokinase activity, enlarges the rate of glycolysis, stimulates glycogen synthesis, increases the storage of glucose as glycogen in the liver and skeletal muscles and decreases glycogen breakdown [14,15].
The influence of insulin on lipid metabolism is in decreasing the rate of lipolysis in adipose tissue and lowering the plasma fatty acid level, stimulation of fatty acid and triacylglycerol synthesis, increasing the uptake of triglycerides (TG) from the blood into adipose tissue and muscles, decreasing the rate of fatty acids oxidation in muscle and liver [14].
With IR insulin-mediated metabolic consequences on lipid and carbohydrate metabolism are changed [15]. The metabolic effect of IR can result in hyperglycemia, hypertension, dyslipidemia, visceral adiposity, elevated inflammatory markers, endothelial dysfunction [13].
Overweight and adiposity, as the result of overnutrition, are the major causative factors that induce the state of lowgrade inflammation due to which accumulation of elevated levels of glucose and/or lipids in blood stream occur that leads to the activation of various transcriptional mediated molecular and metabolic pathways and IR appearance [1]. The factors determining whether or not IR develops as a result of obesity and high body mass index (BMI) are still not completely understood.
The combination of multiple contributing factors (such as puberty, ethnic background, stress etc.) in the obese child along with incapacity of the subcutaneous fat to store excess lipid leading to intrahepatic, intramuscular, and visceral lipid deposition reduce insulin sensitivity and the development of IR and cardiovascular risk factors.
In patients who lost weight and reduce the amount of adipose tissue insulin sensitivity (IS) may return and the association of obese phenotype and type 2 DM incidence disappeare even after adjusting for HOMA-IR [16,17]. Metabolic stress in obesity causes organelle dysfunction, especially the endoplasmic reticulum and mitochondria which regulate the glucose, fat, protein, and cholesterol metabolism. In patients with obesity occur disorders of carbohydrate and fat metabolism [18,19].
The data of frequency of IR diagnosis, the state of carbohydrate and lipid metabolism discharge, and the nature of blood pressure, are debatable and not fully resolved which requires further study. To summarize peculiarities of metabolism and cardiovascular risk in children with IR are actual and known insufficiently.
The purpose of the study was to identify metabolic and somatic changes in children with IR.

Materials and methods
The research study was conducted at the Danylo Halytskiy Lviv National Medical University. Inclusion criteria: IR in children. Exclusion criteria: subjects suffering from infections, endocrine, and genetic disorders, consuming hormones, and suffering from secondary obesity due to other diseases.
Before the clinical assessment, the agreement in participation in the study according to the protocol which was approved by Danylo Halytsky Lviv National Medical University Ethics Commission from parents and children was taken. A detailed medical history with a clinical examination including body mass, body mass index (BMI), neck, waist, and hip circumferences, body surface, and blood pressure (BP) were recorded.
Anthropometry was done according to standard methods. Subjects were classified as generally obese (BMI > 95 %) for sex and age. Abdominal obesity was diagnosed according to more than 90 th percentile of waist circumference (WC) according to age and gender.
Glucose in serum was measured by the glucose oxidative method. Insulin was estimated by enzyme immunoassay DRG Insulin ELISA (Germany). IR was diagnosed according to HOMA-IR. It was determined that HOMA-IR is independently associated with IR, has diagnostic accuracy, and predicts IS [19]. The cut-off values used for the definition of hyperglicemia was 5.6 mM/l, IR for HOMA-IR 2.8, glucose/ insulin ratio 0.48 and hyperinsulinemia 11.5 mcU/ml. Blood pressure (BP) was measured by mechanical manometer three times, the average data were calculated. HDL-C and TG were estimated by the enzymatic method with Cobas Integra HDL-C Test and Cobas Integra TRIGL Test (Roche Diagnostics, Switzerland), respectively.
Data statistical analysis was done by integrative systems for statistical analysis and processing Statistica 10.0 (StatSoft Inc, USA). The overall prevalence of MS and its components were estimated from the data and were associated with 95 % confidence interval. The normality of distribution was estimated according to the Shapiro-Wil-Test Оригінальні дослідження / Original Researches criterion. The results were presented as median with quartile distribution (25 and 75 percentile) and percent of the data in a group. The comparison of groups was done by using Student's T-test and Mann-Whitney U-test depending on parametric distribution of variables. Chi-square test was used for qualitative data presented as positive/negative. For estimation of the link between qualitative characteristics correct Fisher criterion was used. The difference was significant at p-value < .05.
WC in children with IR by 9.2 % was higher than WC in children with NIS (p = .003). Hip circumference in children with IR by 8.0 % was larger than in children with NIS [93.5 (85.0-106.0) cm] (p = .048). Consequently, children and ado lescents with IR were more likely to have higher BMI, WC, hip circumference than children and adolescents with NIS. No significant difference in body weight, height, neck circumference, WC /hip circumference ratio and WC/height ratio, body surface area in children with IR and NIS was observed.
It is natural that the parameters of carbohydrate metabolism in children with IR were significantly different from those in children with NIS (Table 3). Fasted plasma insu-   Оригінальні дослідження / Original Researches lin level in children with IR was 2.3-fold, and fasted plasma glucose level was by 13.0 % higher than those in children with NIS. A significant difference in the HOMA-IR ratio of 2.6fold and the fasted glucose/insulin ratio of 2.1-fold were observed in IR children compared with NIS children. The study of the correlation dependence of anthropometric parameters and carbohydrate metabolism parameters in children with IR revealed a significant association of BMI with fasted plasma glucose (r = .39; p = .023).

Notes: a -difference between the groups is significant with p < .005; b -difference between the groups is significant with p < .0005.
The lipid profile of IR children and NIR children did not differ. Though, in IR children an association of basal glucose with TG/HDL-C ratio (r = .53; p = .001), basal blood insulin with TG (r = .34; p = .018), and TG/HDL-C ratio (r = .54; p = .001) was estimated. The HOMA-IR significantly correlated with VLD-C (r = .40; p = .005), TG (r = .49; p = .001), TG/HDL-C ratio (r = .43; p = .002). The glucose/insulin ratio was in significant association with the TG/non-HDL-C ratio.
Analysis of BP in children with different IS did not reveal the difference between systolic BP (SBP) and diastolic BP (DBP) ( Table 4).
However, the frequency of diagnosis of hypertension in IR children was by 33.8 % higher than in NIS children (p = .001).
It was found that WC in IR children significantly correlated with the SBP (r = .29; p = .042). A highly probable correlation was observed between carbohydrate metabolism and BP: fasting insulin and HOMA-IR significantly correlated with SBP (r = .47; p = 0.001 and r = .53; p = 0.001; respectively) and DBP (r = .43; p = .002 and r =.46; p = .001; respectively). A significantly inverse correlation of the glucose/insulin ratio with SBP was observed (r = -.37; p = .008).
In IR children blood leptin concentration was 1.8 falled higher than the content of leptin in the blood of NIS children (Table 5).

Discussion
Children with IR and NIS did not differ in age and gender, though it was estimated that children with abnormal glucose metabolism were older [20].
The range of IR was 12.7 % in children with normal body mass, 30.7 % in generally obese children, 39.4 % in abdominally obese children. Therefore, with the increase in BMI and the appearance of abdominal obesity the growth of the incidence of carbohydrate metabolism discharge was revealed [20]. It was demonstrated that regardless of the amount of total fat, the accumulation only of abdominal fat and adipose tissue dysfunction lead to an increase in adipokines and the IR appearance [21].
The study of the correlation dependence of anthropometric parameters and carbohydrate metabolism parameters in children with IR revealed a highly probable association of BMI with fasted plasma glucose (r = .39; p = .023). IR was observed to be higher with BMI increasing, which is similar to A. Ahila et al. [22]. The main visual sign in children that there is IR is the presence of general and especially abdominal obesity, in which excess adipose tissue is deposited mainly in the abdomen region. Though, the results demonstrated a notably elevated prevalence rate of IR in overweight and generally obese children [17,23]. The same results were obtained by T. Takahashi et al. [24] that obese/overweight patients had significantly more glucose metabolism disorders than non-obese/non-overweight once. IR is usually one of the first metabolic damage diagnosed in obese children and the key risk factor for the development of comorbidities [25]. It was demonstrated that in patients with weight loss and reduction of the amount of adipose tissue NIS may return even after adjusting for HOMA-IR [16,17].
Normally adipose tissue has a high sensitivity to insulin. In the NIS state, insulin stimulates lipid accumulation through fatty acid uptake, reesterification and lipogenesis and inhibits TG lipolysis [26]. In study of IR children an association of fasting blood glucose and insulin and HOMA-IR with the TG and TG/HDL-C ratio was revealed. That is similar to N. Çin et al. [4] who demonstrated the same positive correlation of HOMA-IR with the TG-glucose index and TG/HDL-C ratio. The Bogalusa Heart Study has shown that IR is closely related to TG levels.
Even compensatory hyperinsulinemia during long course can have a harmful influence on many tissues and, through different mechanisms, contribute to the development of other components of dyslipidemia, such as hypertriglyceridemia [27]. It was established that IR is associated with an increased lipid content and dyslipidemia [28,29]. The obtained data prove that TG, VLDL-C, TG/HDL-C ratio can be used as IR biomarkers.  However, the incidence of hypertension (> 95 th percentile) diagnosis in IR children was 33.8 % higher than in NIS children 2 (p = .001). Insulin plays an important role in the maintenance of vascular homeostasis. Simultaneously insulin stimulates endothelial production of nitric oxide (NO), a crucial vasodilator exerting an antiaggregatory effect that prevents platelet aggregation and leukocyte adhesion to endothelial cells [30] and limiting vascular smooth muscle cells growth and migration, but mediates the release of endothelin 1, known to act as a strong vasoconstrictor. If IR appears, the balance is shifted towards mitogen-activated protein kinase/extracellular signal-regulated kinase (MAPK/ERK), which mediates inflammation, vasoconstriction, and vascular smooth muscle cell proliferation [31]. Therefore, IR coexists with endothelial dysfunction and hypertension [32].

Parameters
The pathogenesis of hypertension associated even with compensatory hyperinsulinemia is stimulation in vascular smooth muscle and endothelial cells increased production of endothelin, PAI-1, proinflammatory cytokines, and an augmented surface expression of adhesion molecules [27]. IR and hyperinsulinemia cause a number of hemodynamic changes that contribute to increased BP and the development of hypertension. An association was found between fasting insulin levels and subsequent increases in BP. The main role in the development and progression of IR and related metabolic disorders is played by adipose tissue of the abdominal area, neurohormonal disorders that accompany abdominal obesity, and increased activity of the sympathetic nervous system.
In the IR children's blood leptin concentration was 1.8 fallen higher than the content of leptin in the blood of NIS children, so with increasing blood insulin levels and the development of IR observed a proportional increase in leptinemia and the development of leptin resistance [33]. It was found that leptin mediates insulin secretion and sensitivity in peripheral tissues. Hyperleptinemia reflecting leptin resistance plays an important role in the development of IR making leptin its possible biomarker [34].
High correlation of leptin with basal insulin and HOMA-IR are in agreement with the findings of the studies done by P. Das et al. [35]. It was demonstrated that there is a simultaneous occurrence of IR and leptin resistance in obesity. The correlations of leptin levels with insulin, and HOMA-IR levels surely indicate the functional link between these two hormones. Leptin has an inhibitory effect on the insulin secretion, which is due to the leptin-induced proinflammatory cytokines such as C-reactive protein and Interleukin-6, causing apoptosis of pancreatic β-cells [34,36].
It was demonstrated that leptin acts as a signal from adipocytes to pancreatic cells and reports IS. Leptin can be involved in the induction of IR by cross-reacting with insulin receptor substrates. Also, macrophage infiltration of adipose tissue and high levels of leptin in obesity lead to increased production of IL and TNF, which prevent the effects of insulin on IS tissues, and increase IR. Imbalance and dysregulation of adipokine production are involved in metabolic disorders of obesity [37].
In IR children was found an essential correlation of blood leptin with WC, which was in agreement with the findings of the studies done by J. Mohiti et al. stated that it was obesity leading to increased insulin and leptin levels, resulting in IR and leptin resistance [38]. Leptin levels in general increase proportionately with body fat mass.
Leptin may be the marker of obesity and IR at the same time [34], and conversely obesity is the major pathogenic factor common to both leptin resistance and IR.

Conclusions
IR is related to cardiometabolic risks, such as general and abdominal obesity, hypertension, dyslipidemia, hyperleptinemia, and leptin resistance.
IR is a screening marker for children and adolescents with an increased risk of cardiometabolic diseases.
Consolidation of the actions of pediatricians and endocrinologists, new methods of early diagnostics, and improvement of the system of IR prevention are necessary.