Abstract
CONTEXT: Thyroid-stimulating hormone (TSH) levels physiologically increase with age. Applying a fixed adult reference range to older adults may cause overdiagnosis of subclinical hypothyroidism (SCH). This study examined whether adjusting the upper limit of the TSH reference range is appropriate by evaluating reference ranges and risks of geriatric conditions in older adults. DESIGN AND SETTING: We analyzed 1626 community-dwelling adults aged 65-84 years from the Bunkyo Health Study. Reference ranges for TSH, free triiodothyronine (FT3), and free thyroxine (FT4) were calculated using the 2.5th-97.5th percentiles. Participants with normal FT4 were classified into 3 groups by TSH: euthyroid (0.61-4.23 mIU/L), mild SCH (>4.23 to <7.0 mIU/L), and moderate-to-severe SCH (≥7.0 mIU/L). Associations between TSH categories and geriatric conditions were evaluated using logistic regression. RESULTS: With increasing age, TSH tended to increase, while FT3 decreased. The 2.5th-97.5th percentile range of TSH in the total cohort was 0.49-5.56 mIU/L. After adjusting for confounders, moderate-to-severe SCH was significantly associated with cerebral microbleeds (odds ratio [OR]: 4.22, 95% CI: 1.49-11.97) and sarcopenia (OR: 3.27, 95% CI: 1.02-10.52) compared to the euthyroid group. Mild SCH was not significantly associated with any geriatric condition. CONCLUSION: Age-related TSH elevation is common among older adults. Mild SCH was not linked to geriatric conditions, whereas moderate-to-severe SCH showed associations with cerebral microbleeds and sarcopenia. Given the limited number of participants in this category, these findings should be interpreted cautiously and require confirmation in longitudinal studies.