Thyroid-associated ophthalmopathy with diplopia: a magnetic resonance imaging–based correlational study of intraorbital tissues and their response to therapy

To investigate the correlation of extraocular muscle and intraorbital fat involvement with diplopia and their responsiveness to intravenous glucocorticoid therapy in patients with thyroid-associated ophthalmopathy (TAO) using magnetic resonance imaging (MRI). Clinical records, laboratory indices, and imaging parameters were retrospectively collected from 195 patients with TAO (114 with diplopia and 81 without) and 30 contemporaneous patients with normal orbital MRI findings. Parameters, including extraocular muscle thickness, extraocular muscle volume, signal intensity ratio of the extraocular muscle (M-SIR), intraorbital fat volume (FV), SIR of intraorbital fat (F-SIR), and orbital volume (OV), were measured. Univariate and multivariate logistic regression analyses were used to determine independent predictors of TAO-related diplopia and therapeutic efficacy. The diagnostic performance of these indicators was assessed using receiver operating characteristic curves. In total, 450 orbits (228 with diplopia, 162 without diplopia, and 60 normal controls) were analyzed. Most parameters were significantly higher in patients with TAO and diplopia compared with those without diplopia. However, stepwise multivariate logistic regression identified inferior rectus thickness, M-SIR mean , FV/OV, and F-SIR mean as the independent risk factors with the highest diagnostic value for TAO-related diplopia. The combined index of these indicators showed superior diagnostic performance compared with individual parameters. Similarly, M-SIR mean and F-SIR max were identified as independent determinants of therapeutic responsiveness for diplopia using stepwise multivariate logistic regression. TAO-related diplopia and its therapeutic outcomes are associated with extraocular muscle changes and intraorbital fat involvement. A comprehensive analysis of both structures can provide more objective evidence for clinical decision-making.