Effects of Thyroid Status on Regional Brain Volumes

A Diagnostic and Genetic Imaging Study in UK Biobank

Tom Chambers; Richard Anney; Peter N. Taylor; Alexander Teumer; Robin P. Peeters; Marco Medici; Xavier Caseras; D. Aled Rees


J Clin Endocrinol Metab. 2021;106(3):688-696. 

In This Article

Abstract and Introduction


Background: Thyroid hormone is essential for optimal human neurodevelopment and may modify the risk of attention-deficit/hyperactivity disorder (ADHD). However, the brain structures involved are unknown and it is unclear if the adult brain is also susceptible to changes in thyroid status.

Methods: We used International Classification of Disease-10 codes, polygenic thyroid scores at different thresholds of association with thyroid traits (P T-values), and image-derived phenotypes in UK Biobank (n = 18 825) to investigate the effects of a recorded diagnosis of thyroid disease and genetic risk for thyroid status on cerebellar and subcortical gray matter volume. Regional genetic pleiotropy between thyroid status and ADHD was explored using the GWAS-pairwise method.

Results: A recorded diagnosis of hypothyroidism (n = 419) was associated with significant reductions in total cerebellar and pallidum gray matter volumes (β [95% CI] = −0.14[−0.23, −0.06], P = 0.0005 and β [95%CI] = −0.12 [−0.20, −0.04], P = 0.0042, respectively), mediated in part by increases in body mass index. While we found no evidence for total cerebellar volume alterations with increased polygenic scores for any thyroid trait, opposing influences of increased polygenic scores for hypo- and hyperthyroidism were found in the pallidum (P T < 1e−3: β [95% CI] = −0.02 [−0.03, −0.01], P = 0.0003 and P T < 1e−7: β [95% CI] = 0.02 [0.01, 0.03], P = 0.0003, respectively). Neither hypo- nor hyperthyroidism showed evidence of regional genetic pleiotropy with ADHD.

Conclusions: Thyroid status affects gray matter volume in adults, particularly at the level of the cerebellum and pallidum, with potential implications for the regulation of motor, cognitive, and affective function.


Thyroid hormone (TH) is essential for optimal neurodevelopment, exemplified by the severe intellectual disability that affected patients with untreated congenital hypothyroidism prior to the introduction of neonatal screening. TH influences several aspects of neurodevelopment, including differentiation and proliferation of neuronal precursors, neuronal migration, and myelination.[1] Early pregnancy may be a critical window, since suboptimal maternal gestational thyroid function has been implicated as a risk factor for neurodevelopmental disorders,[2,3] while behavioral difficulties may affect children born to mothers exposed to excess thyroxine (T4) in pregnancy.[4]

The brain regions involved in any potential effects of TH, however, are unclear, although the cerebellum is emerging as a major target. Cerebellar ontogenesis is profoundly influenced by TH: in rodents, hypo- and hyperthyroidism affect foliation, differentiation, and migration of granule cells; Purkinje cell arborization; and neuronal cell death.[1] Such effects might provide a mechanistic link to the risk of neurodevelopmental disorders, since the cerebellum and subcortical connections are implicated as core structures in the pathogenesis of ADHD.[5]

However, human structural brain studies in thyroid disease are limited. In a prospective study of >1900 mother-child pairs from the Generation R cohort, an inverse U-shaped association was noted between maternal free T4 levels and offspring gray matter volume,[6] suggesting that both low and high TH levels may adversely affect human neurodevelopment. Altered thyroid status may also affect gray matter volume in adults,[7,8] although the numbers studied to date are small.

A new resource for relating recorded diagnoses and genetics to neuroimaging is UK Biobank. This prospective epidemiological study of 500 000 middle-aged volunteers[9] includes structural brain magnetic resonance imaging (MRI) of regional gray matter volumes. We thus set out to investigate whether a recorded diagnosis of hypo- or hyperthyroidism in UK Biobank was associated with alterations in gray matter volume in adults, hypothesizing that any effects observed would be opposing, and most apparent in the cerebellum. We additionally sought to explore whether genetic risk for thyroid disorders had similar effects on gray matter volume, and whether such a risk overlapped with that for ADHD.