Endocrine Abstracts

Objective: Adequate thyroid hormone availability is essential for foetal neurodevelopment. While overt maternal hypothyroidism is associated with lower offspring IQ, recent studies suggested an inverse U-shaped association, with suboptimal offspring neurodevelopment for both the lower and higher thyroid function range. Our aim is to investigate the association of gestational thyroid function with offspring IQ scores, assess if this association is different based on the gestational age at blood draw or TPOAb positivity.

Methods: We performed an individual participant data meta-analysis using data from prospective birth cohorts on gestational thyroid function and offspring IQ. Exclusion criteria were pre-existing thyroid disease, thyroid hormone altering medication usage, multiple gestation. IQ measurements <50 or >150 were winsorized. Neurodevelopmental tests were standardized to a mean of 100 and standard deviation of 15 to facilitate comparison among cohorts. We performed hierarchical linear mixed regression models accounting for repeated measurements, adjusting for potential confounders including maternal age, education, ethnicity, body mass index, smoking status, parity, gestational age at blood sampling and fetal sex. We used multilevel multiple imputation for missing covariate data and inverse probability weighting to account for attrition.

Results: The final study population comprised 14, 767 participants from 11 cohorts of whom 6, 064 had repeated IQ measurements (median age of measurement 6.1 years, interquartile range [IQR] 3.1-8.7 years). There was an inverted U-shaped association of FT4 concentrations with child IQ (P=0.00048). For both low and high FT4 this corresponded with a mean difference in IQ scores of -5.5 to -6.5 in the full range and -1.0 to -1.5 in the euthyroid range. A leave-one-out analyses to prevent a single cohort from driving the association yielded similar results. After stratification for gestational age at blood sampling (median 13.2 weeks, IQR 11.6-17.5), there was no association between FT4 and child IQ scores after 25 weeks of gestation, although data were limited in this period. There was no statistically significant association of TSH, TPOAb positivity, subclinical hypothyroidism, isolated hypothyroxinaemia, subclinical hyperthyroidism or overt hyperthyroidism with child IQ scores.

Conclusions: Both low and high maternal FT4 concentrations, but not TSH, in pregnancy are associated with lower child IQ scores. This is the first multicentre study to replicate these findings and add important data on time dependency and persistency of the effect until older offspring age. These results emphasize that caution should be exercised when supplementing thyroxine in pregnancy at the risk of overtreatment and possible detrimental effects on offspring neurodevelopment.