Histogram of Apparent Diffusion Coefficient to Evaluate the Activity of Thyroid-Associated Ophthalmopathy

This study aimed to evaluate the activity of extraocular muscles (EOMs) in patients with thyroid-associated ophthalmopathy (TAO) using turbo spin echo imaging. By analyzing tissue heterogeneity, apparent diffusion coefficient (ADC) histogram analysis offers enhanced insights into edema within the EOMs.

Eighty-eight patients with TAO were retrospectively evaluated and allocated into active (n = 24, clinical activity score [CAS] ≥ 3) and inactive (n = 64, CAS < 3) groups. The parameter values of the ADC histogram of EOMs were measured; the efficacy of ADC histograms in distinguishing between TAO activity and inactivity was assessed using receiver operating characteristic curves. Multifactorial logistic regression was used to determine active TAO predictors.

The minimum, maximum, median, mean; and 1st, 5th, 10th, 25th, 75th, 90th, 95th, and 99th percentiles of the ADC histograms were higher in patients with active than that in participants with inactive TAO. The area under the curve (AUC) of the 10th percentile of the ADC histogram and the median distinguishing between active and inactive TAOs were both 0.791 (both p < 0.05), and the AUCs of the combined model of age, sex, smoking, and the 10th percentile in the ADC histogram were better than those of their individual models and the combined model of age, sex, and smoking (all p < 0.05). Smoking and male sex, along with the median > 1.26 μm²/s, entropy > 4.03, and standard deviation (SD) > 0.4 of the ADC histogram, were significant predictors of TAO activity, with odds ratios of 2.741 and 6.806, 5.070, 2.652, and 2.197, respectively (all p < 0.05).

ADC histograms provide a new method for distinguishing active from inactive TAO, and the 10th percentile enhances the clinical diagnosis of active TAO. In addition to male sex and smoking, an ADC histogram median > 1.26 μm²/s, entropy > 4.03, or SD > 0.4 may also predict active TAO.