Serum Thyroid-Stimulating Hormone as a predictor of cognitive impairment in an elderly cohort

Background: It is unclear whether in late life serum thyroid-stimulating hormone (TSH) predicts risk of developing cognitive impairment. Objective: This study investigated the prospective relationship of serum TSH with the risk of developing mild cognitive impairment (MCI), Alzheimer's disease (AD) and vascular dementia (VaD) in an elderly cohort with a 4-year follow-up. Methods: Data are for 660 subjects aged 65 years and older from an Italian population-based cohort who were cognitively normal at an extensive assessment in 1999/2000 and underwent follow-up assessment in 2003/2004. Serum TSH was measured at baseline. Multinomial logistic models adjusted for sociodemographic and cardiovascular risk factors were used to investigate the association of serum TSH (both as a tertile and continuous log-transformed variable) with risk of incident MCI, AD and VaD diagnosed according to international criteria. Results: Over 3.8 +/- 0.7 years of follow-up, there were 149 incident MCI cases (77 with impairment of memory and 72 with impairment of nonmemory domains) and 86 incident dementia cases (53 with AD, 28 with VaD). No association between baseline TSH and risk of developing any MCI subtype or AD was found. The highest TSH tertile had a threefold higher increased risk of VaD (OR: 3.25, 95% CI: 1.01-10.77, p = 0.048) compared to the lowest tertile. Risk of VaD increased about 60% for each 1 SD increase in log-transformed TSH (OR: 1.61, 95% CI: 1.06-2.44, p = 0.025). Conclusions: In this elderly cohort, baseline TSH was not related to the risk of developing MCI or AD, but high TSH was associated with an increased risk of VaD. These results suggest further need for research using larger samples to examine the role of TSH as a predictor of VaD and the role of thyroid autoimmunity in vascular cognitive impairment.