Longitudinal Associations Between Ambient Air Pollution With Insulin Sensitivity, β-Cell Function, and Adiposity in Los Angeles Latino Children

Abstract

Evidence suggests that ambient air pollution (AAP) exposure may contribute to the development of obesity and type 2 diabetes. The objective of this study was to determine whether exposure to elevated concentrations of nitrogen dioxide (NO₂) and particulate matter with aerodynamic diameter <2.5 (PM_2.5) had adverse effects on longitudinal measures of insulin sensitivity (S_I), β-cell function, and obesity in children at high risk for developing diabetes. Overweight and obese Latino children (8-15 years; n = 314) were enrolled between 2001 and 2012 from Los Angeles, CA, and followed for an average of 3.4 years (SD 3.1 years). Linear mixed-effects models were fitted to assess relationships between AAP exposure and outcomes after adjusting for covariates including body fat percent. Higher NO₂ and PM_2.5 were associated with a faster decline in S_I and a lower S_I at age 18 years, independent of adiposity. NO₂ exposure negatively affected β-cell function, evidenced by a faster decline in disposition index (DI) and a lower DI at age 18 years. Higher NO₂ and PM_2.5 exposures over follow-up were also associated with a higher BMI at age 18 years. AAP exposure may contribute to development of type 2 diabetes through direct effects on S_I and β-cell function.

Figure 1

Effects of long-term average AAP exposure during follow-up on metabolic indices from an FSIVGTT at age 18 years in Latino children. The estimated percent difference in S_I (A), AIR_g (B), and DI (C) at age 18 years for a 5-ppb and 4 μg/m³ difference in long-term NO₂ and PM_2.5 exposure, respectively. Results are displayed as the estimated percent difference with 95% CIs. Each model was adjusted for sex, Tanner stage, season of testing (warm/cold), social position, body fat percent, study wave, and study entry year. *P < 0.05.

Figure 2

Effects of long-term average AAP exposure during follow-up on fasting and 2-h insulin and glucose measures from an OGTT at age 18 years in Latino children. The estimated percent difference with 95% CIs are shown for fasting insulin (A), fasting glucose (B), 2-h insulin (C), and 2-h glucose (D) at age 18 years for a 5-ppb and 4 μg/m³ difference in long-term NO₂ and PM_2.5 exposure, respectively. Each model was adjusted for sex, Tanner stage, season of testing (warm/cold), social position, body fat percent, study wave, and study entry year. *P < 0.05; **P < 0.01.

Figure 3

Effects of long-term AAP exposure during follow-up on BMI and adiposity at age 18 years in Latino children. The estimated difference with 95% CIs are shown for BMI (A), body fat percent (B), SAAT (C), and IAAT (D) at age 18 years for a 5-ppb and 4 μg/m³ difference in long-term NO₂ and PM_2.5 exposure, respectively. Each model was adjusted for sex, Tanner stage, season of testing (warm/cold), social position, body fat percent (where appropriate), study wave, and study entry year. *P < 0.05; **P < 0.01; #P = 0.05.