Insulin Sensitivity and Associated Risk Factors in Mexican Children and Adolescents

One hundred male and female children and youth subjects aged between 10 and 20 years, from different socioeconomic levels, provided general information about their health status as well as blood samples for a 75-g oral glucose tolerance test (2) and basal and 2-h insulin and lipid determinations (triglycerides and HDL, LDL, and total cholesterol) (3,4). Insulin sensitivity was estimated by the insulin sensitivity index (ISI)_0,120 method (5,6). Acanthosis nigricans (7) and blood pressure measurements were performed by a trained physician. Anthropometrical measurements included body weight, height, BMI, and waist circumference. Body composition and fat distribution were made by body density from air-displacement plethysmography measurements (8,9) and by dual-energy X-ray absorptiometry for abdominal fat (10). Subjects completed a questionnaire regarding their family history of type 2 diabetes. The data were analyzed using the statistical package NCSS 2001 (Number Cruncher Statistical System, Windows version; Kaysville, UT). Comparative analysis was made for the general lineal model, adjusting by confounding variables and expressed as means ± SE. Associations were explored by simple and multiple linear regression.

There were no diabetic subjects detected, and only 2% showed impaired glucose tolerance; however, important risk factors for type 2 diabetes were found in this group of youths. Table 1 shows insulin sensitivity status and variables related to the metabolic syndrome. Body weight, BMI, Z score of BMI for age, percent body fat, abdominal fat, waist circumference, serum triglycerides, and systolic blood pressure were higher (P < 0.0007 to P < 0.0001) in subjects with low insulin sensitivity (lower tertile of ISI_0,120), and HDL cholesterol was lower (P < 0.02). Also, multiple regression showed that the best predictors of insulin sensitivity (ISI_0.120) were abdominal fat (P < 0.00001), serum triglycerides (P < 0.014), and family history of type 2 diabetes (P < 0.016). Acanthosis nigricans was present in 33% of the subjects and correlated positively with 2-h glucose, fasting insulin, 2-h insulin, serum triglycerides, systolic and diastolic blood pressure, BMI, waist circumference, percent body fat, and abdominal fat (P < 0.01 to P < 0.00001) and negatively with HDL cholesterol and ISI_0,120 (P < 0.0001). Abdominal fat (in kilograms) showed stronger correlation coefficients than waist circumference with variables like 2-h glucose, fasting and 2-h insulin, ISI_0,120, HDL cholesterol, and serum triglycerides.

In this study, no diabetic subjects were detected, and only two subjects showed impaired glucose tolerance. Insulin sensitivity status showed that subjects in the lower tertile had higher body weight, total body fat, higher abdominal fat, lower HDL cholesterol, and higher blood pressure than those in the two upper tertiles. Most of these are major features in the insulin resistance syndrome. In Hispanic subjects living in the U.S., greater insulin resistance has been observed compared with other minorities and is partly attributed to greater adiposity (11). The best predictors of insulin sensitivity were abdominal fat, serum triglycerides, and family history of type 2 diabetes, which is very similar to studies reported by Cruz et al. (12) in obese Hispanic children aged 8–13 years, where family history for type 2 diabetes and visceral fat were independently and negatively associated to insulin sensitivity. Thirty-three percent of the subjects showed positive for Acanthosis nigricans upon examination of the neck, and this was associated with variables related to the metabolic syndrome, similar to what other studies in youths have reported (13), where 10- to 12-year-old children had lower HDL, higher LDL, elevated triglycerides, higher total cholesterol/HDL, and a higher prevalence of overweight/obesity when Acanthosis nigricans was present. Most studies (14–18), as in our case, find that being positive for Acanthosis nigricans is related to hyperinsulinemia or indicators of insulin resistance in adults and children. In populations of emerging economies like Mexico, type 2 diabetes and cardiovascular disease are substantially increasing (19) and are accompanied by changes in lifestyle that include diet and physical inactivity. Children and adolescents are no exception to these changes. Despite not finding subjects with type 2 diabetes in this study and having a low proportion of subjects with impaired glucose tolerance, there were important features of the metabolic syndrome that were strongly associated with insulin sensitivity and potential factors that could lead to type 2 diabetes and cardiovascular disease later in life. It is important to identify these early metabolic disturbances that can eventually result in type 2 diabetes in overweight and obese children. Even though a low prevalence of type 2 diabetes was found in the Mexican children and adolescents of this study, metabolic disturbances related to insulin sensitivity, such as alteration of lipid levels, body and abdominal fat, blood pressure, and the presence of Acanthosis nigricans, are features to be monitored. This is why the clinician should consider some of these risk indicators for early detection and the prevention of type 2 diabetes in youths as well as for potential intervention programs at the community level.

The project was partially financed by the International Atomic Energy Agency, Vienna, Austria (CRP 10602/RO).

The authors thank the participating children, adolescents, and their parents. We thank Bertha I. Pacheco for technical assistance and Dr. Silvia Y. Moya and Dr. Luis Quihui for manuscript review.