Changes in Thyroid Function and Autoimmunity in Older Individuals: Longitudinal Analysis of the Whickham Cohort

Abstract

Background: Longitudinal studies of thyroid function have demonstrated differing results. It remains unclear whether changes in thyroid function affect the diagnosis of subclinical thyroid dysfunction with aging.

Methods: Survivors of the Whickham cohort study were evaluated on 2 occasions between the years 2008 and 2012 and 2016 and 2019. Serum TSH, free T4 (FT4), free T3 (FT3), and thyroid peroxidase antibody (TPOAb) were measured on both occasions using the same assay under similar conditions. Individuals with known thyroid disease or on medications affecting thyroid function were excluded. Comorbidities were noted, functional mobility was assessed by the timed up-and-go test, and muscle function was evaluated by the hand grip strength test.

Results: In 204 individuals (mean age 77.0 [±6.6] years, 114 [56%] female), followed over a median (interquartile range) of 7.8 (7.3-8.2) years, serum TSH increased by 0.29 mU/L (12.4%), FT3 and TPOAb reduced by 0.1 pmol/L (-2.1%) and 0.6 U/L (-11.2%), and there were no significant changes in FT4 levels. The calculated upper limit of serum TSH increased over the follow-up period from 4.74 mU/L to 6.28 mU/L. The relationship between serum TSH and FT4 at both time points was not significantly different. Utilizing standard laboratory reference ranges, the prevalence of subclinical hypothyroidism increased from 3.5% at baseline to 9.0% at follow-up. However, adopting a visit-specific TSH reference range reduced the prevalence of subclinical hypothyroidism at both time points to 2.0%.

Discussion: Thyroid function demonstrates subtle but significant changes with age. Utilizing standard reference ranges tends to increase the diagnosis of subclinical hypothyroidism in older euthyroid individuals. Our data suggest that adopting age-appropriate TSH reference ranges may reduce the risk of diagnosing and (potentially unnecessarily) treating subclinical hypothyroidism.

Keywords: TSH; aging; thyroid function.

Figure 1.

Longitudinal change in thyroid function and TPOAb over time. The line in the box represents the median value; the lower edge of the box represents the lower quartile and the upper edge of the box shows the upper quartile. The values at which the horizontal lines stop are the values of the upper and lower values of the data. The single points on the diagram show the outliers. The P-values were obtained from paired t-test analyses (TSH and TPOAb were log-transformed prior to analysis). Abbreviations: TPOAb, thyroid peroxidase antibody.

Figure 2.

The relationship between longitudinal change in TSH with baseline age. Scatterplot with fitted regression line. The P-value was derived from a multivariable linear regression analysis that included baseline age, sex, smoking status, body mass index, Charlson Comorbidity Index, baseline thyroid peroxidase antibody levels, and delta free FT4 or delta free T3.

Figure 3.

The relationship between serum TSH and free T4 levels at each visit. The slopes of the regression line assessing the relationship between TSH and free T4 at both time points were not statistically different.

Figure 4.

Change in serum TSH levels in individuals with subclinical hypothyroidism and their baseline thyroid peroxidase antibody status.