Endocrine Abstracts

Background: Previous genetic and animal studies implied a potential association between thyroid hormone and osteoarthritis (OA), but this has not been confirmed in the general population. We aim to investigate whether thyroid function is associated with hand, hip or knee OA.

Methods: We included 9,054 participants from Rotterdam Study with baseline measurement of thyroid-stimulating hormone (TSH), free thyroxine (FT4) and radiographs. Joint radiographs were scored using the Kellgren and Lawrence (KL) score system, with OA defined as KL ≥2. Overall progression of knee and hip OA was defined as any increase of KL score during follow-up except the increase from 0 to 1. Severity of OA was determined by the sum of KL scores. We used multivariable regression models to investigate the association of thyroid function with prevalence, severity, incidence and progression of OA. The associations were adjusted for age, sex and additionally adjusted for body mass index (BMI) and physical activity. We conducted stratified analyses by sex, age and weight and post-hoc analyses stratified by weight-bearing physical activity.

Results: Higher levels of FT4 were associated with an increased risk of prevalent knee OA (age- and sex-adjusted odds ratio [OR] 1.02, 95% CI 1.00-1.04). The effect estimate became stronger with further adjustment of BMI and physical activity (OR 1.04, 95% CI 1.01-1.06), corresponding to an OR of 1.62 (95% CI 1.21-2.18) across the reference range of FT4. We identified a consistent positive association of FT4 with severity of knee OA. In longitudinal analyses, there was a borderline significant association between FT4 and progression of knee OA (age- and sex-adjusted OR 1.02, 95% CI 0.99-1.05). The association became stronger after adjusting for BMI and physical activity (OR 1.03, 95% CI 1.00-1.07), and restricting to euthyroid participants (OR 1.05, 95% CI 1.02-1.08). However, we did not identify any significant association of TSH and FT4 with prevalence, severity, incidence or progression of hand and hip OA. Stratified analysis indicated that the association of FT4 with prevalent knee OA was more pronounced among individuals aged ≥65 years, and those with BMI ≥30 kg/m². In post-hoc analyses, participants with high levels of weight-bearing physical activity showed a significant association between FT4 and prevalent knee OA (OR 1.05, 95% CI 1.01-1.10).

Conclusion: Our study indicated that higher FT4 levels may increase the risk of knee OA, particularly in individuals with extra joint-loading, such as obesity and weight-bearing physical activity. Future studies are warranted to validate our findings.