Objectives: The objective of this study is to investigate whether body mass index (BMI) changes are associated with fasting glucose and insulin resistance (IR) in early childhood.

Methods: From the Environment and Development of Children (EDC) cohort, 334 children who visited at ages 2 and 4 were included in this study. Height and weight were measured at ages 2 and 4, and fasting glucose and insulin were assessed at age 4. Homeostatic model assessment of insulin resistance (HOMA-IR) was calculated as insulin (μIU/ml) × glucose (mg/dl)/405. The BMI Z-score [BMI (Z)] quartiles for each age group were defined as Q4, ≥75th percentile; Q2-3, 25th to 75th percentile; and Q1, <25th percentile. Glucose, insulin, and the HOMA-IR were compared between groups according to the change in BMI (Z) from age 2 to 4.

Results: Children who stayed in Q4 at both ages had higher fasting glucose (92.2 vs. 88.0 and 87.1 mg/dl), insulin (3.2 vs. 2.5 and 2.3 μIU/ml), and HOMA-IR (0.68 vs. 0.54 and 0.52) than children who stayed in Q1 or Q2-3 (all P<0.01). Children in Q4 at both ages had higher fasting glucose than children whose BMI (Z) increased from Q1 or Q2-3 to Q4 (92.2 vs. 87.3, P<0.001). The BMI (Z) category at age 2 of children who were in Q2-3 at age 4 did not affect glucose or IR at 4 years.

Conclusion: The group of children within the highest BMI (Z) quartile at both 2 and 4 years of age had higher fasting glucose and IR at age 4 than any other BMI (Z) change group.