Disparities of adiposity explained ethnic differences in insulin sensitivity and secretion traits in adolescents with normoglycaemia.

AIMS: Excessive adiposity contributes to impaired insulin sensitivity and insulin secretion among different ethnic groups. We aim to investigate whether disparities in adiposity can explain the ethnic differences in insulin sensitivity and secretion traits within a multi-ethnic cohort of normoglycaemic adolescents.

METHODS

A total of 3,349 adolescents with normoglycaemia (mean age: 11.3 ± 1.2 years) from four ethnic groups (Asians, Non-Hispanic Blacks, Hispanics, and Non-Hispanic Whites) were recruited from ten international centres between 2013 and 2016. Insulin sensitivity and secretion traits (insulin sensitivity, insulin secretion and the balance between their relationship) were evaluated from oral glucose tolerance test. Adiposity was assessed using body mass index (BMI), waist-to-height ratio (WHtR), and fat mass index (FMI). Adjusted partial R-squared values were used to quantify the variance of insulin sensitivity-secretion traits explained by each adiposity measure, after controlling for age, sex, maternal gestational diabetes, Tanner stage, and field centre. Differences of estimated marginal means (ΔEMM) were utilized to compare the differences of insulin sensitivity and secretion traits between Non-Hispanic Whites and other ethnic groups.

RESULTS

Adiposity measures explained the highest proportion of variance of insulin sensitivity and secretion traits in Hispanics (31.97% - 43.56%), but the lowest in Non-Hispanic Whites (18.09% - 24.80%), after controlling for covariates. Disparities in BMI could fully explain the differences in insulin sensitivity as well as the balance between sensitivity and secretion in the comparison between Non-Hispanic Blacks and Non-Hispanic Whites after controlling for covariates (pΔEMM > 0.05), whilst disparities in adiposity could only partially explain the observed ethnic differences in insulin sensitivity and secretion traits in other comparisons across ethnic groups (pΔEMM < 0.05). The contribution of each adiposity measure to these differences varied across ethnic groups, with FMI, BMI and WHtR being the most relevant measures for Asians, Non-Hispanic Blacks and Hispanics respectively, compared with Non-Hispanic Whites.

CONCLUSIONS

Disparities in adiposity could partially or fully explain the observed ethnic differences in insulin sensitivity and secretion traits. Our study provides new understandings regarding the contribution of adiposity in determining insulin sensitivity and secretion traits, and may provide novel insights on how adiposity may shape ethnic differences in metabolic profiles.