Predicting glucocorticoid effectiveness in thyroid eye disease: combined value from serological lipid metabolism and an orbital MRI parameter

Abstract

Purpose The aim was to determine the combined value of serological lipid metabolism and an orbital MRI quantitative parameter in predicting the effectiveness of glucocorticoid (GC) therapy in patients with thyroid eye disease (TED). Methods This study retrospectively enrolled 46 patients with active and moderate-to-severe TED (GC-effective group, n = 29; GC-ineffective group, n = 17). Serological lipid metabolism, the orbital MRI-based minimum signal intensity ratio of extraocular muscles (EOM-SIRmin), as well as other clinical parameters before GC therapy were collected and compared between the two groups. Multivariate logistic regression and receiver operating characteristic curve analysis were adopted to identify independent predictable variables and assess their predictive performances. Results Compared to the GC-ineffective group, the GC-effective group showed lower serum total cholesterol levels (P = 0.006), lower serum low-density lipoprotein cholesterol levels (P = 0.019), higher EOM-SIRmin values (P = 0.005), and shorter disease durations (P = 0.017). Serum total cholesterol and EOM-SIRmin were found to be independent predictors of GC-effective TED through multivariate analysis (odds ratios = 0.253 and 2.036 per 0.1 units, respectively) (both P < 0.05). The integration of serum total cholesterol ≤4.8 mmol/L and EOM-SIRmin ≥ 1.12 had a better predictive efficacy (area under the curve, 0.834) than EOM-SIRmin alone, with a sensitivity of 75.9% and a specificity of 82.4% (P = 0.031). Conclusion Serological lipid metabolism, combined with an orbital MRI-derived parameter, was a useful marker for predicting the effectiveness of GCs in patients with active and moderate-to-severe TED.