Predictive model for the progression of inactive thyroid eye disease: a retrospective study.

Abstract

This study aimed to develop a model for predicting the progression of inactive thyroid eye disease (TED). We retrospectively analyzed 107 patients with inactive TED, who did not receive any TED treatment during the follow-up period of 3-6 months. A multivariable regression analysis was employed to examine the clinical risk factors influencing the progress of TED and a predictive model based on risk factors was established. Seventeen (15.9%) patients had progressed at the last observation. Male (p = 0.024), having smoking history (p = 0.009), compound indicator combined the duration of TED and immunosuppressive treatment history before 3 months (p = 0.026) and more severe diplopia (p = 0.015) were found to be the potential risk factors for progression. Among the subset of patients who exhibited progression, there was a notable escalation in the severity of proptosis (p = 0.021), margin reflex distance-1 (p = 0.031), and clinical activity score (p = 0.011). The multivariable stepwise regression analysis identified positive smoking history as well as untreated and having a duration of TED > 12 months as independent risk factors of TED progression. A predictive model including these risk factors was built, with the AUC (Area Under Curve) score of 0.783. The predictors of TED progression include smoking history and composite indicators of duration time and treatment history. Our model may provide recommendations for immunosuppressive therapy and prognostic predictions for inactive TED patients preparing for surgery.