Surrogate Biomarkers of Insulin Resistance in Obese Children and Adolescents

Abstract

Obesity is the most common reason for the insulin resistance (IR) though obesity in children is not always associated with IR. The markers based on fasting glucose and/or insulin levels and alternative parameters with conventional lipid-based indexes are used in assessing the IR in clinical practice. The purpose of the research was to assess the relationship between alternative IR surrogate biomarkers as follows: triglycerides (TG) to high-density lipoprotein cholesterol (HDL-C), TG/HDL-C ratio; single-point insulin sensitivity estimator (SPISE) index; triglyceride-glucose index (TyG), with generally accepted indices based on the insulin/glucose ratios, such as HOMA-IR, QUICKI, Caro, in a cohort of obese children and adolescents. Materials and methods used: a single-stage multicenter comparative study was conducted involving 127 patients aged 6 to 17 y/o with simple obesity of varying degrees. The following indices were calculated: HOMA-IR = fasting insulin (mU/ml) x fasting plasma glucose (FPG) (mmol/L) / 22.5; QUICKI = 1/Log (Fasting Insulin, µU/ml) + Log (Fasting Glucose, mg/dl); Caro = fasting plasma glucose (mmol/l) / fasting insulin (μU/ml); TG/HDL-C = TG (mmol/L) / HDL-C (mmol/L); triglyceride-glucose index TyG = Ln [fasting TG (mg/dl) x fasting plasma glucose (mg/dl)/2], where Ln is the logarithm; SPISE = 600 × HDL-C0.185/(TG0.2 x BMI1.338). Results: 75 (59%) boys and 52 (41%) girls were included in the study; the median age was 12.9 [10.4; 14.9] y/o. Pre-pubertal development was recorded in 25 (19.7%), pubertal - in 102 (80.3%). The prevalence of impaired glucose tolerance (IGT) was 5.5%. In children with IGT, the SPISE index was lower than in children with normal glucose tolerance (3.69 [3.49; 4.37] vs. 4.8 [3.99; 5.55], p=0.021) with no statistically significant differences between these groups in other markers. In the IGT subgroup, a high negative correlation was found between SDS BMI and SPISE (R=-0.97). In the general SDS group, BMI correlated with insulin (R=+0.24), TG (R=+0.2), HOMA (R=+0.24), Caro (R=-0.26), TG/HDL-C (R=+0.28) and SPISE (R=-0.65). The SPISE index was lower in girls than in boys (4.94 [3.63; 5.73] vs. 4.07 [3.59; 5.16], p = 0.045). No statistically significant differences were recorded for other gender-related markers. A negative correlation was also found between the obesity degree and SPISE in the subgroups of girls (R=-0.49) and boys (R=-0.70) as well as among pre-pubertal children of both genders (R=-0.61). The maximum correlation between SDS BMI and SPISE was recorded in the subgroup of teenagers of both genders (R=-0.88). In the general group, TyG was weakly correlated with HOMA-IR (R=+0.26) and QUICKI (R=-0.24), while SPISE showed a weak association with all studied insulin indices; no statistically significant correlation between insulin indices and TG/HDL-C was found. In the subgroup of girls, no significant relationships between the evaluated markers were found, in the subgroup of boys, TyG weakly correlated with HOMA-IR (R=+0.29) and QUICKI (R=-0.32). Stronger correlations were recorded in boys between SPISE and all studied insulin indices. In the subgroup of prepubertal patients, a correlation was found between TyG and QUICKI (R=-0.44), SPISE and Caro (R=+0.56), SPISE and HOMA-IR (R=-0.50). The TG/HDL-C ratio did not statistically significantly correlate with traditional IR markers. TyG showed weak correlations with HOMA-IR and QUICKI in the general group and the boys’ subgroup. In the prepubertal period, TyG only weakly correlated with QUICKI and did not show strong associations with insulin-based indices in pubertal children. The SPISE index showed correlations with insulin markers both in the main group and in individual subgroups. Conclusion: the SPISE index can be considered as a relatively simple and accessible method for assessing IR in obese children and adolescents.