INTRODUCTION

The prevalence of childhood obesity continues to grow worldwide¹. Increased sugar consumption is considered to be a contributor to the global epidemics of obesity and diabetes and their associated cardiometabolic risks². Overconsumption of added sugar is associated with obesity, non-alcoholic fatty liver disease (NAFLD), and insulin resistance (IR)³-⁵. Some of these associations may be the result of hepatic de novo lipogenesis stimulation by dietary sugars⁶,⁷. Among added sugars, fructose is emerging as a potentially harmful component⁷,⁸. The intake of large amounts of fructose creates an unregulated source of carbon precursors for hepatic lipogenesis⁹, which contributes to the development of NAFLD¹⁰ and hepatic IR⁹,¹¹. Hepatic IR is characterized by an inability of insulin to suppress glycogenesis and glycogenolysis, and it is closely associated with a dysfunction of the adipose tissue¹². The liver has been shown to be the main site of production of the insulin-like growth factor binding protein-1 (IGFBP-1) in humans, which has been proposed as a hepatic insulin-resistance marker¹³,¹⁴. IGFBP-1 is unique among other hepatic insulin-resistance markers because it is regulated by insulin and predicts glucose tolerance worsening; moreover, it is less labile than insulin¹³,¹⁵. Since IGFBP-1 is inversely correlated with liver fat, it is independent of obesity. Hepatic fat accumulation is tightly linked to hepatic IR and is characterized by a decreased ability of insulin to suppress hepatic glucose and triglyceride (TG)-rich, very-low-density-lipoprotein particle production in the liver¹⁴. On the other hand, one of the weaknesses of IGFBP-1 is that its reproducibility in serum, such as other indicators of IR, has been reported to decline with deteriorating glucose tolerance¹³. Previous research has shown that IGFPB-1 promotes the regeneration of β cells¹⁶ and that high levels of IGFBP-1 are associated with a reduction in diabetes risk¹⁵,¹⁷. Low serum IGFPB-1 levels have also been observed in obesity, hyperinsulinemic conditions¹⁸,¹⁹, and NAFLD conditions²⁰. A previous study in children with obesity in our group showed that 73.1% had a high homeostatic model assessment of IR (HOMA-IR) and low IGFBP-1 levels, which could be an indication of whole-body IR and hepatic IR²¹.

Few studies have been conducted in children, and to the best of our knowledge, there are no studies evaluating hepatic IR, NAFLD, and fructose consumption in children with obesity. Therefore, the aim of this study was to evaluate IGFBP-1 (used as a marker of hepatic IR), NAFLD and their relationship with fructose consumption in children with obesity.

METHODS

RESULTS

A total of 83 children were studied with a mean age of 9.2 ± 1.3 years; 48.2% were male and 51.8% were female. About 93% of the girls presented IR, and 72.1% had a TG/HDL-C index > 2. The main contributor to fructose intake was the sugar-sweetened beverages group, with 42.75 ± 19.72 g/day; it was followed by the sugar-sweetened foods group, with 14.56 ± 9.39 g/day; the fruit group, with 4.69 ± 5.52 g/day; and finally, the vegetables and natural juices groups, with < 1.0 g/day.

Table 1 shows a comparison by gender for general and metabolic characteristics; the girls showed greater waist circumference (p = 0.002), TGs (p = 0.009), VLDL (p = 0.01), HOMA-IR (p = 0.00001), and TG/HDL-C index (p = 0.012), and they had lower IGFBP-1 levels (p = 0.0001). No significant differences were found in fructose intake from the food groups analyzed (results not shown).

Table 2 shows the comparison of metabolic variables according to low or high IGFBP-1 levels. The group with low IGFPB-1 was older (p = 0.008) and had higher BMI (p = 0.00002) and waist circumference (p < 0.0001), as well as higher HOMA-IR (p < 0.0001). In relation to dietary variables, a higher energy consumption (p = 0.012), carbohydrates (p = 0.051) and total fructose intake (p = 0.022) were found in the group with low IGFPB-1. No significant differences were found in fructose intake from the food groups analyzed (results not shown).

In the total group, 81% had NAFLD in its different levels (light 34% and moderate/severe 47%). The group with moderate/severe NAFLD had higher BMI (p = 0.009), waist circumference (p = 0.046), insulin (p = 0.022), HOMA-IR (p = 0.036), carbohydrates intake (p = 0.005), total fructose intake (p = 0.001) (Fig. 1), and higher fructose intake from sugar sweetened beverages (p = 0.007). In this group, we also found lower levels of IGFBP-1, but these were not statistically significant (Table 3).

Figure 1 Comparison of fructose consumption according to the non-alcoholic fatty liver disease (NAFLD) degree. (1): none; (2): light; (3): moderate and severe NAFLD. NAFLD: non-alcoholic fatty liver disease. 

Negative correlations were found between IGFBP-1 and BMI (r = −0.42; p = 0.02), HOMA-IR (r = −0.61; p = 0.002), and total fructose intake (r= −0.25; p = 0.03).

The multiple regression analysis is presented in Table 4: IGFBP-1 was found to be associated with BMI, HOMA-IR, and total fructose intake (R² = 0.41; F = 10.75 p < 0.0001), and this association persisted after adjustment by gender (R² = 0.11; F = 10.75 p < 0.0001).

DISCUSSION

In Mexico, obesity is an important public health problem. The National Health and Nutrition Survey 2016 reported that the combined prevalence of overweight and obesity in children between 5 and 11 years was 33.2%. The obesity prevalence in girls was 12.2% and 18% in boys. In the past decades, high fructose intake has been associated with increased risk of obesity development and the consequent metabolic and inflammatory diseases³,³². Liver function is particularly vulnerable to an increasing fructose intake since this organ is responsible for approximately 90% of the total metabolism of this sugar³³. The liver has also been found to be the main site of production of IGFBP-1 in humans, a marker of hepatic IR. However, few studies related to these markers have been conducted in children and none, to the best of our knowledge that evaluates hepatic IR, NAFLD, and fructose consumption in children with obesity. In this study, we evaluated school-age children and found that in the group with low levels of IGFBP-1, the children showed higher levels of IR and HOMA-IR, and IGFBP-1 showed a negative association with HOMA-IR. These results suggest that higher hepatic IR is present with higher whole IR. Similar results have been found by other authors in elderly men¹⁸ and nondiabetic patients on peritoneal dialysis³⁴. An explanation for these results may be found in recent studies that proposed that IGFBP-1 promotes the regeneration of β-cells¹⁵, re-establishing the pancreatic function while the total IR persists or worsens. Reports from animal studies indicate that obesity is associated with the suppression of the hepatic expression of mRNA of IGFBP-1 and a marked reduction of total concentrations of circulating IGFBP-1³⁵.

In our study, the group with lower levels of IGFBP-1 had a higher consumption of carbohydrates and total fructose, and in the multivariate analysis, IGFBP-1 showed a negative association with fructose consumption. In this sense, the liver is the primary site of de novo lipogenesis, the process by which fatty acids are synthesized from dietary precursors, predominantly carbohydrates³⁶. Excessive fructose consumption may also have significant effects on lipid metabolism, contributing both to steatosis and to increased circulating TG levels by means of the inhibition of the enzyme carnitine palmitoyltransferase 1A². Changes in IGFBPs have been correlated with some metabolic syndrome (Mets) components, and predominantly with obesity³⁷,³⁸. Reinehr et al.²² found that IGFBP-1 correlated significantly with most of the Mets components (waist circumference, TGs, and HOMA-IR). Similar results were found in our group of children with obesity.

In our study group, 81% of the children showed NAFLD at levels (light 34%, and moderate/severe 47%) higher than those found in previous studies in our group (42.5%)²¹. These children also had higher BMI and larger waist circumference; similar results were found in children from the general population and studies based on child obesity clinics, although the prevalence was lower³⁹. The children with moderate/severe NAFLD showed higher HOMA-IR, which could be explained by reports that hyperinsulinemia leads to de novo lipogenesis, which is partially responsible for the accumulation of TGs in the hepatocytes and the development of steatosis⁴⁰. However, the mechanisms by which high-fructose feeding cause’s hyperinsulinemia and IR remain uncertain2. Fructose-induced steatosis may contribute to hepatic IR through increased hepatic diacylglycerol accumulation, protein kinase C activation, and impairment of insulin-mediated Akt2 activation⁴¹,⁴². However, whether steatosis itself can cause hepatic IR remains controversial⁴³,⁴⁴. In our study, higher consumption of carbohydrates and fructose observed in the group with moderate/severe NAFLD was similar to other studies⁶,¹⁰,⁴⁵. High carbohydrates intake exceeding the energetic requirements and above the liver storage capacity in the form of glycogen, are converted into fatty acids to be stored as triacylglycerol, leading to NAFLD⁴⁶. On the other hand, a meta-analysis indicated that consumption of fructose has more damaging effects on the hepatic insulin sensitivity than does the isocaloric consumption of carbohydrates⁴.

Likewise, Gugliucci⁴⁷ state that approximately 90% of fructose intake is metabolized in the liver, while a large part of the glucose consumed passes through the liver and finds its way to skeletal muscle, where it degrades to CO₂, H₂O, and adenosine triphosphate, and to the adipose tissue, where it is converted into phosphate glycerol for the synthesis of TGs and production of energy. Therefore, it is undeniable that in this multi-factorial pathology of NAFLD, elevated consumption of carbohydrates, especially fructose, is a major risk factor for its development.

A previous study in children by our group showed that the restriction of high-fructose foods with a decrease in caloric and carbohydrate intake, at 6 weeks induced a decrease of TG levels and hepatic steatosis⁴⁸. In the present study, the group of children without steatosis and the group with mild steatosis had similar fructose intake; besides, they had similar metabolic and anthropometric characteristics. An explanation to these similarities could be that ultrasound is not sensitive enough to detect mild states of steatosis⁴⁹.

It is important to highlight that our results show that girls have higher levels of TGs, HOMA-IR, I-TG/HDL, and lower IGFBP-1; all of them considered cardiovascular risk factors. A possible explanation is that girls had higher fructose consumption, which contributed to the development of hepatic IR, total IR, and higher TGs which could put them at higher risk of developing cardiometabolic diseases in early adulthood; thus, it is important to evaluate all of these risk factors from an early age.

This study had several weaknesses. First, the small sample size and that only 68 samples of IGFBP-1 were processed due to insufficient serum; also, the transversal nature of the study which did not allow us to evaluate causality; and finally, the use of ultrasound as a diagnostic tool for steatosis. Ultrasound detects fatty liver disease in patients with moderate and high-fat content but is limited in patients with low-fat content⁴⁹. Furthermore, conventional ultrasound signs may be operator-dependent and subjective, although the intra- and inter-observer agreement in our study was good to excellent.

The group with lower levels of IGFBP-1 (marker of hepatic IR) was associated with BMI, HOMA-IR, and fructose consumption after adjustment by gender; the group with higher severity of NAFLD showed higher HOMA-IR and fructose consumption, and the metabolic alterations were more evident in females (higher TGs, HOMA-IR, and lower IGFBP-1). These results support the importance to evaluate children to prevent the development of chronic diseases.