Abstract
Background Selenium (Se) status may influence activity and treatment response in thyroid eye disease (TED), but evidence from Se‑sufficient regions is limited. We aimed to evaluate the association between serum Se levels and clinical and serological outcomes in patients with TED treated with intravenous steroid therapy and Se supplementation. Serum Se increased from 125.05 to 167.75 µg/L throughout the follow‑up period (p < 0.001). Baseline Se correlated inversely with Clinical Activity Score (CAS) change at 2 months (ρ=-0.308; p = 0.047). A baseline Se threshold of 147.53 µg/L showed modest discrimination for ≥ 2-point CAS reduction (area under the curve [AUC] 0.653; sensitivity 50.0%; specificity 85.7%; positive predictive value 63.6%; negative predictive value 77.4%). Patients with Se ≥ 147.53 µg/L showed greater CAS reduction than those with lower levels (median - 2.0 vs. - 1.0, p = 0.012), and retained lower baseline‑adjusted final CAS in a rank‑based analysis of covariance (p = 0.017). Se levels were not significantly correlated with thyroid-stimulating hormone receptor antibody or thyroid-stimulating immunoglobulin changes. In a Se‑replete cohort receiving intravenous steroid plus Se, higher pretreatment Se was associated with greater CAS improvement. Given the modest AUC and limited power (59.5%), these hypothesis‑generating findings warrant prospective validation.