Endocrine Abstracts

Objectives: To assess antenatal and individual-level predictors of thyroid function across two longitudinal cohorts of children.

Methods: Data were taken from two cohorts i) CATSII: 5-10 year-old children enrolled in the CATS (Controlled Antenatal Thyroid Screening) study II from the UK ii) ‘PAX8’: Children from the ENID (Early Nutrition and Immune Development) trial, a pregnancy and infant supplementation trial in The Gambia (Africa) recruited to the PAX8 study aged 5-8 years based on PAX8 gene region methylation at age 2. CATSII children were classified as ‘Normal’ if their mothers had normal thyroid function during pregnancy, ‘Untreated’ in case of maternal suboptimal gestational thyroid function (SGTF) and ‘Treated’ in case of maternal SGTF treated with levothyroxine during pregnancy. PAX8 participants were recruited into “high” and “low” groups i.e. from the top and bottom methylation quantiles. Children with positive TPO antibodies or with abnormal thyroid function were excluded from analysis.

Results: Data were available for 80 CATS children (excluded: n =3 positive TPO antibodies, n =1 high TSH) and 118 from PAX8 study (none positive for TPO antibodies). For both cohorts: In unadjusted comparisons, girls had higher FT4 compared to boys (CATS: 15.1 pmol/L vs. 14.37 pmol/L, P < 0.05. PAX8: 14.0 pmol/L vs. 13.3 pmol/L, P < 0.05) and girls had higher FT3 compared to boys in CATSII cohort (5.9 pmol/L vs. 5.4 pmol/L, P < 0.01). There were no significant associations with TSH. For the CATSII cohort: In multiple linear regression models, adjusted for age and BMI, Sex (Male β=-0.47, P = 0.16) was not significantly associated with FT4. However, ‘Normal’ group children had significantly lower FT4 (β=-0.93, P < 0.05) compared to ‘Untreated’ group. In multiple linear regression models, FT3 was significantly associated with Sex (Male β=-0.39, P < 0.01), Age (β=-0.13, P < 0.05) and BMI (β=0.05, P < 0.05) but not in the Treated group. For the PAX8 cohort: In multiple linear regression models, adjusted for age and BMI, Sex (Male β=-0.78, P < 0.01) and PAX8 methylation group (Low PAX8 β=0.80, P < 0.01) were significantly associated with FT4. In multiple linear regression models, FT3 was significantly associated with Age (β=-0.18, P < 0.01) but not Sex, BMI or PAX8 methylation group.

Conclusions: Untreated SGTF was associated with higher FT4 in offspring compared to those whose mothers had normal gestational thyroid function and points to an early programming effect on children’s FT4. Lower PAX8 methylation was associated with higher FT4. Children’s Sex, Age, BMI were associated with a range of thyroid measures with some consistencies across disparate cohorts.