Introduction
Overt hypothyroidism, with elevated thyrotropin (TSH) and low thyroid hormone (TH), is a rare but reversible cause of cognitive dysfunction, often screened for during dementia evaluations.1 Such testing identifies what is often called “subclinical hypothyroidism” (elevated TSH and normal TH) in 15–20% of older adults.2 Isolated TSH elevation has not been associated with significant cognitive effects,3 and thyroid hormone treatment trials in older adults have not shown benefit for cognitive symptoms4 or endpoints.5 Many studies,6 but not all,3 have found the opposite: higher incident cognitive dysfunction with lower TSH and higher TH. Importantly, low TSH is often iatrogenic,7 which may account for some heterogeneity in the literature. For example, TSH in the Health, Aging, and Body Composition Study (HealthABC) predicted incident dementia when exogenous thyrotoxicosis was included.8
We hypothesize that aggressive case finding in those with early signs of cognitive dysfunction leads to detection and treatment of isolated TSH elevation, increasing the risk of iatrogenic thyrotoxicosis, and potentially accelerating cognitive decline. With millions of individuals older than 70 years living with various forms of dementia, an understanding of clinical practice patterns and the need to individualize the approach to abnormal TSH elevation in older adults with dementia is critical to optimizing care.
Methods
We assessed the relationship between TSH level, thyroid hormone use, and dementia diagnosis in HealthABC, a longitudinal study of 3,075 healthy volunteers aged 70–79 living independently in Memphis, Tennessee and Pittsburgh, Pennsylvania from 1997–1998.8 Briefly, all participants had baseline Modified Mini-Mental Status scores ≥78. TSH was measured in Year 2, with free T4 measured for a TSH <0.45mIU/L or >7.0mIU/L. Medication review and dementia diagnosis were performed annually. Our study included Year 2 visits with a measured TSH not on anti-thyroid medications (n=2,798). None were missing dementia status, seven (0.25%) were missing medication information. Thyroid function categorizes included: euthyroid (TSH; 0.45 – 4.49mIU/L), high TSH (TSH >4.5mIU/L), low TSH (TSH <0.45mIU/L). Multinomial logistic regression was performed using STATA® (16.0 IC, College Station, TX).
Results
Dementia prevalence in Year 2, was 2.3% (62/2,798) and did not differ by race, sex, smoking status or site (data not shown). Those with dementia were older (76 vs. 75 years, p<0.01) and had a lower BMI (26 vs. 27, p=0.04) than those without dementia. There was a notable, but non-significant, higher percentage of thyroid hormone use with a dementia diagnosis (12.9% vs. 9.8%, p=0.4). Dementia was associated with a lower likelihood of high TSH and a higher likelihood of low TSH (4.8% vs. 11.7% and 8.1% vs 3.2% respectively, p <0.04), as illustrated in Figure 1. None of the 23 participants with overt hypothyroidism had a dementia diagnosis.
Figure 1: Distribution of TSH by Thyroid Hormone Use and Dementia Status.
TSH distribution shifts depending on dementia diagnosis and the use of thyroid hormone. The proportion of both high TSH (light blue) and low TSH (purple) is impacted by dementia status and thyroid hormone supplementation. The reference population of untreated participants with dementia (A) is compared to those without dementia on thyroid hormone (B), those with dementia not on thyroid hormone (C) and those with dementia on thyroid hormone treatment (D).
In adjusted models, thyroid hormone supplementation was associated with increased risk of both over-treatment (RRR=9.9, 95%CI 6.2–15.6) and under-treatment (RRR=2.4, 95%CI 1.7–3.3; Table 1). Each year of age increased the risk of having a low TSH by 19% compared to being euthyroid (RRR=1.2, 95%CI 1.0–1.2). Women were at greater risk of both low TSH (RRR=1.9, 95%CI 1.2–3.1) and high TSH (RRR=1.3, 95%CI 1.0–1.6) compared to men.
Table 1:
Risk of Low or High TSH With Thyroid Hormone Use and Dementia
| Covariate | Low TSH (TSH < 0.45mIU/L) | High TSH (TSH > 4.5mIU/L) | ||
|---|---|---|---|---|
| Model 1 RRR (95% CI) |
Model 2 RRR (95% CI) |
Model 1 RRR (95% CI) |
Model 2 RRR (95% CI) |
|
| Without Interaction | ||||
| Dementia | 2.3 (0.8–6.1) | 2.1 (0.8–5.8) | 0.4 (0.1–1.3) | 0.4 (0.1–1.3) |
| Thyroid hormone use | 10.7 (6.9–16.5)* | 9.9 (6.2–15.6)* | 2.7 (1.9–3.7)* | 2.4 (1.7–3.3) |
| Interaction Model | ||||
| Dementia only | 1.8 (0.4–7.4) | 1.5 (0.3–6.2) | 0.2 (0.02–1.2) | 0.2 (0.02–1.2) |
| Thyroid hormone use only | 10.4 (6.6–16.2)** | 9.4 (5.9–15.1)** | 2.6 (1.9–3.6)** | 2.3 (1.6–3.2) |
| Dementia + Thyroid hormone use | 44.8 (8.8–227.5)** | 47.2 (8.9–249.0)** | 5.4 (0.9–32.6) | 4.7 (0.8–28.3) |
| Age | 1.2 (1.0 – 1.2)# | 1.0 (0.98 – 1.1) | ||
| Female | 1.9 (1.2 – 3.1)* | 1.3 (1.0 – 1.6)# | ||
| Black | 1.3 (0.8 – 2.0) | 0.5 (0.4 – 0.7)** | ||
| BMI | 0.96 (0.9–1.0) | 0.99 (0.97–1.0) | ||
P<0.001
p<0.01
p<0.05.
Model 1: dementia and thyroid hormone use.
Model 2: dementia, thyroid hormone use, age, sex, race, BMI.
Dementia modified the relationship between thyroid hormone use and TSH: the likelihood of a low TSH among treated participants with dementia was 47.2 times (95%CI 8.9–249.0) that of untreated participants without dementia (Table 1).
Discussion
Our findings raise three concerns. First, despite few dementia cases diagnosed by Year 2, when TSH is measured, we found a striking 90% lower prevalence of untreated TSH elevation among dementia compared to non-dementia participants (Figure 1). The rate of 10.7% in non-dementia participants is expected from the literature.9 Therefore, the very low rate (1.9%) in those with dementia supports our hypothesis that systematic case-finding leads to treatment initiation for isolated elevated TSH. Alternatives are less likely, as a protective effect of untreated high TSH has not been seen in prior studies.3,6 This finding also impacts interpretation of prospective research studies in which the presence of confounding by indication is a potential limitation when thyroid hormone use is an exclusion criterion.
Secondly, we observe higher rates of over-treatment in those with dementia, imprecisely measured here and needing confirmation in larger studies. This could result from challenges with medication adherence or dose escalation for ongoing symptoms, both of which risks harm from thyrotoxicosis in this vulnerable population.
Finally, if low TSH and high TH levels are risk factors for dementia progression,6 and this status often results from thyroid hormone use,7 prescribing supplementation for emerging cognitive dysfunction has the potential to exacerbate cognitive decline. An emerging consensus that in older adults isolated elevated TSH can represent stress responses rather than disease,10 reinforces our recommendation that clinicians be mindful of the physiology when interpreting TSH during dementia evaluations.
Acknowledgments
Sponsor’s Role: Sponsors did not contribute to the development of the study and paper.
Footnotes
Conflict of Interest: The authors declare that the research was conducted without any commercial or financial relationships that could be construed as a potential conflict of interest.
Disclosures:
EJA, JM and JSRM: NIA R01AG064256; EO: NIA R01AG057725; EMS: NIA Intramural Research Program: JSRM: Turock Family Foundation;
References:
- 1.Knopman DS, DeKosky ST, Cummings JL, et al. Practice parameter: diagnosis of dementia (an evidence-based review). Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology. 2001;56(9):1143–1153. [DOI] [PubMed] [Google Scholar]
- 2.Hollowell JG, Staehling NW, Flanders WD, et al. Serum TSH, T(4), and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III). The Journal of clinical endocrinology and metabolism. 2002;87(2):489–499. [DOI] [PubMed] [Google Scholar]
- 3.van Vliet NA, van Heemst D, Almeida OP, et al. Association of Thyroid Dysfunction With Cognitive Function: An Individual Participant Data Analysis. JAMA Intern Med. 2021;181(11):1440–1450. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Mooijaart SP, Du Puy RS, Stott DJ, et al. Association Between Levothyroxine Treatment and Thyroid-Related Symptoms Among Adults Aged 80 Years and Older With Subclinical Hypothyroidism. JAMA. 2019;322(20):1977–1986. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Parle J, Roberts L, Wilson S, et al. A randomized controlled trial of the effect of thyroxine replacement on cognitive function in community-living elderly subjects with subclinical hypothyroidism: the Birmingham Elderly Thyroid study. The Journal of clinical endocrinology and metabolism. 2010;95(8):3623–3632. [DOI] [PubMed] [Google Scholar]
- 6.Rieben C, Segna D, da Costa BR, et al. Subclinical Thyroid Dysfunction and the Risk of Cognitive Decline: a Meta-Analysis of Prospective Cohort Studies. The Journal of clinical endocrinology and metabolism. 2016;101(12):4945–4954. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Mammen JS, McGready J, Oxman R, Chia CW, Ladenson PW, Simonsick EM. Thyroid hormone therapy and risk of thyrotoxicosis in community-resident older adults: Findings from the Baltimore Longitudinal Study of Aging. Thyroid : official journal of the American Thyroid Association. 2015;25(9):979–986. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Aubert CE, Bauer DC, da Costa BR, et al. The association between subclinical thyroid dysfunction and dementia: The Health, Aging and Body Composition (Health ABC) Study. Clinical endocrinology. 2017;87(5):617–626. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Canaris GJ, Manowitz NR, Mayor G, Ridgway EC. The Colorado thyroid disease prevalence study. Archives of Internal Medicine. 2000;160(4):526–534. [DOI] [PubMed] [Google Scholar]
- 10.Mammen JS. Interpreting Elevated TSH in Older Adults. Curr Opin Endocr Metab Res. 2019;5:68–73. [DOI] [PMC free article] [PubMed] [Google Scholar]