Assessment of insulin sensitivity from measurements in fasting state and during an oral glucose tolerance test in obese children.

BACKGROUND: Few previous studies have examined the validity of the fasting glucose-to-insulin ratio (FGIR), homeostasis model assessment of insulin resistance (HOMA-IR) and quantitative insulin-sensitivity check index (QUICKI) in pediatric populations.
OBJECTIVE: To compare simple indices of insulin resistance calculated from fasting glucose and insulin levels with insulin sensitivity indices (area under the response curve [AUCinsulin], insulin sensitivity index [ISI-compositeL) determined by oral glucose tolerance testing (OGTT) in obese children.
METHODS: One hundred and forty-eight obese children and adolescents (86 girls and 62 boys, mean age: 10.86 +/- 3.08 years, mean body mass index (BMI): 27.7 +/- 4.2) participated in the study. OGTT was performed in all participants. After glucose and insulin measurements from OGTT, the children were divided into two groups according to the presence or absence of insulin resistance. Insulin sensitivity indices obtained from the OGTT were compared between the groups. The total plasma glucose response and insulin secretion were evaluated from the AUC estimated by the trapezoid rule. Cut-off points, and sensitivity and specificity calculations were based on insulin resistance with receiver operating characteristic curve (ROC) analysis.
RESULTS: The prevalence of insulin resistance, glucose intolerance and dyslipidemia was 37.1%, 24.3% and 54% in obese children, respectively. The groups consisted of 93 children without insulin resistance (54 girls and 39 boys; mean age: 10.5 +/- 3.3 years; mean BMI: 27.0 +/- 4.2) and 55 children with insulin resistance (32 girls and 23 boys; mean age: 11.4 +/- 2.5 years; mean BMI: 27.9 +/- 3.9). There were significant differences in mean FGIR (10.0 +/- 7.2 vs 5.6 +/- 2.8, p < 0.001), HOMA-IR (3.2 +/- 2.3 vs 4.9 +/- 2.3, p < 0.001) and QUICKI (0.33 +/- 0.03 vs 0.30 +/- 0.02, p < 0.001) between the groups. The cut-off points for diagnosis of insulin resistance were < 5.6 for FGIR (sensitivity 61.8, specificity 76.3), > 2.7 for HOMA-IR (sensitivity 80, specificity 59.1), and < 0.328 for QUICKI (sensitivity 80, specificity 60.2).
CONCLUSIONS: Indices derived from fasting samples for diagnosis of insulin sensitivity are reliable criteria in obese children and adolescents. HOMA-IR and QUICKI appeared to have similar sensitivity and specificity and to have higher sensitivity than FGIR.