Application of Multiparameter Quantitative Magnetic Resonance Imaging in the Evaluation of Graves' Ophthalmopathy

Background Assessment of the activity of Graves' ophthalmopathy (GO) is difficult. Existing methods need improvement. Purpose Investigate the application of multiparametric magnetic resonance imaging (MRI) in GO. Study Type Retrospective. Population A total of 235 GO patients (age: 38.8 ± 13.4 years; 90 male; 96 active patients). Field Strength/Sequence Short‐tau inversion recovery (STIR) fast spin echo, multiecho spin echo T2 mapping and 3D T1‐weighted fast field echo sequences at 3.0 T. Assessment Two physicians assessed the mean and maximum signal intensity ratio of extraocular muscles to white matter (SIR), T2 relaxation time (T2RT), extraocular muscle area (EMA), fat fraction (FF), retrobulbar fat volume (RFV), and extraocular muscle volume (EMV). Clinical activity score (CAS) ≧ 3 was in active stage. Statistical Tests The optimal cut‐off point of diagnostic efficacy was selected using receiver operating characteristic (ROC) curve analysis and evaluated using area under the curve (AUC), compared using Student's t test, analysis of variance or Kruskal–Wallis H test. The correlation used Pearson correlation analysis. The discriminant equation used a binary logistic regression analysis. P < 0.05 was considered statistically significant. Results The SIRmean, SIRmax, T2RTmean, T2RTmax, EMA, and EMV in active GO patients were significantly higher than those in inactive and were positively correlated with CAS ( r = 0.276, 0.228, 0.438, 0.388, 0.502, and 0333, respectively). The FFmax of active patients was significantly lower than that of inactive patients and was negatively correlated with CAS ( r = −0.44). Logistic regression analysis indicated that T2RTmean was independently associated with GO active periods and had good diagnostic performance (area under ROC curve = 0.736, sensitivity 70.7%, specificity 69.3%). T2RTmean ≧ 74.295 could be a diagnostic cut‐off for judging GO activity (sensitivity 55.3%). Conclusion SIR, T2RT, EMV, and FF can quantitatively assess the activity and severity of GO and can potentially provide a basis for clinical judgment and selection of treatment options. Evidence Level 4. Technical Efficacy Stage 2.