Endocrine Abstracts (2018) 59 OC2.3 | DOI: 10.1530/endoabs.59.OC2.3

Prospective serum thyroid function and cognitive decline in the very old: the Newcastle 85+ study

Earn Gan1,2, Mohammad Yadegarfar3, Carol Jagger3 & Simon Pearce1,2


1Institute of Genetic Medicine, Newcastle University, Newcastle upon Tyne, UK; 2Royal Victoria Infirmary, Newcastle upon Tyne, UK; 3Institute of Health & Society, Newcastle University, Newcastle upon Tyne, UK.


Context: Perturbations in thyroid function are common in older people, and subclinical hyperthyroidism has been associated with increased risk of dementia in people aged 55 years and above. The significance of subtle perturbations of thyroid function in the very old remains poorly understood.

Objective: This study sought to determine if subtle abnormalities of thyrotropin and variations of free thyroid hormones within the reference range correlate with cognitive impairment in the very old, using data from the Newcastle 85+ study.

Design: A cohort of 85-year-old individuals was assessed for their health status and thyroid function. Cross-sectional and prospective data (up to 5 years follow-up) were analysed using linear mixed and regression models for global and memory-specific cognitive performance in relation to baseline and 3-year changes in serum thyrotropin (TSH), free T4 (FT4) and free T3 (FT3).

Setting and participants: Six hundred and forty-two 85-year-olds with TSH ranged between 0.1–10 mU/l, normal FT3/FT4 levels and who were not taking thyroid-interfering medication were included.

Results: After adjusting for age, sex, years of education and smoking, cognition (MMSE and memory sensitivity index) was associated with baseline log-transformed TSH (P=0.012) and free T3 (P<0.01). After additional adjustment for potential confounders, including depression, physical activity and chronic disease status, both baseline log-TSH and FT3 remained significantly associated with global cognition (P<0.05), with lower log-TSH and FT3 correlated with worse cognitive outcome. Reduction in TSH over the initial 3 years increased the odds of cognitive impairment at 60-month (OR 1.60 (95% confidence interval 1.14–2.24), P=0.006).

Conclusions: Individuals aged 85 years with low but unsuppressed TSH or low normal free T3 had a significantly worse cognition at baseline. We show, for the first time, that a decreasing TSH trajectory anticipates the development of cognitive decline in later life.

Article tools

My recent searches

No recent searches.