Multi-modal Medical Images for Extraocular Muscle Fibrosis Assessment

Thyroid eye disease (TED) is an autoimmune orbitopathy characterized by inflammation and remodeling of the orbital tissues. These changes can lead to the development of symptoms such as proptosis, diplopia, and cosmetically disturbing. Strabismic diplopia in the active phase TED results from restriction of muscle movement due to inflammatory and oedema of extraocular muscles (EOMs). Recent studies have demonstrated that patients with active TED receiving teprotumumab experienced a meaningful improvement in proptosis and strabismic diplopia. However, there are still patients with strabismic diplopia who require strabismus surgery in the inactive phase TED. Unfortunately, the dose-response relationship in strabismus surgery for TED still exhibits significant variability, potentially leading to unsatisfactory postoperative alignment.

EOMs are manipulated mechanically during surgery for strabismus. In the progression of TED, the heterogeneity in the biomechanical properties of the EOMs caused by fibrosis may contribute to the unpredictability of strabismic diplopia. Thus, evaluating their biomechanical property changes has the potential to provide invaluable information that improves our understanding of the pathogenic mechanisms of TED-related strabismus. Real-time tissue elastography (RTE) is a noninvasive modality that provides information on periocular and ocular tissues elasticity. It could also account for strain distribution inhomogeneity across large tissue areas, potentially serving as a powerful tool for detecting and monitoring pathological processes affecting EOM stiffness.

Biomechanical property changes are often based on corresponding changes in the morphology. Orbital magnetic resonance imaging (MRI) could help to quantitatively assess the clinical activity, fibrosis and other pathological changes of TED. Studies recently reported that native T1 mapping values provide the best discrimination of morphological and fibrotic information of EOMs, and the extracellular volume (ECV) fraction is a reliable tool to quantify EOMs fibrosis. Nevertheless, the correlation of this method to relevant muscle elasticity architectures remains unavailable.

Methods Study Design and Participants Consecutive patients (n=40) with moderate-to-severe TED were retrospectively enrolled from Tianjin Eye Hospital between 2020 and 2024. All patients underwent MR with T1 mapping and RTE. The patients were categorized into either the active or inactive TED group based on the Clinical Activity Score (CAS). The sex-matched control group (n=18) comprised healthy individuals and patients with orbital symptoms but normal findings on orbital MRI. All participants underwent a comprehensive ophthalmic examination, including slit-lamp biomicroscopy, Hertel exophthalmometry and intraocular pressure. A blood sample was obtained on the day of MR scanning to measure hematocrit. Participants who reported any of the following were excluded: (1) any ophthalmopathies (e.g., glaucoma, diabetic retinopathy and optic nerve disorders); (2) patients with dysthyroid optic neuropathy (DON). Our institutional ethics committee approved all study protocols, and patients provided written informed consent.

MR image acquisition and image analysis All participants were examined using a 3.0-T MR scanner (Siemens Magnetom Verio dot, Siemens Healthcare) and a 64-channel brain coil. On the basis of a previous report, the scan sequences included routine sequences (T1-weighted sequence, T2-weighted sequence, fat-suppressed, STIR sequence) and pre/postcontrast T1 mapping using a dual-flip-angle method. Postcontrast T1 mapping was repeated in identical prescription as precontrast T1 mapping 15 min after contrast injection. The extra-cellular volume (ECV) was calculated following the previously described methodology. The region of interest (ROI) was drawn at the thickest part of the muscle and its adjacent upper and lower levels in selected axial planes, and then MRI parameters of the medial rectus (MR) and lateral rectus (LR) muscles were measured from both sides. Two independent radiologists (L.X. and J.M.) delineated ROIs on the EOMs of all participants. Three repeated measurements were performed for each muscle slice, and the mean value was used for analysis. Radiologist L.X. repeated all evaluations twice at one-month intervals to assess intraobserver reliability.

Measurement of extraocular muscles stiffness The elasticity of the extraocular muscles was assessed using an ultrasound scanner equipped with real-time tissue elastography (EUB-7500; Hitachi Medical Systems, Tokyo, Japan). The images were acquired and analyzed following the previously described methodology. The system displays real-time color-coded elastography images overlaid on B-mode ultrasound. As this provides only qualitative stiffness assessment, we used the elastic ratio method for quantitative analysis. ROI was simultaneously placed on extraocular muscles and orbital fat, and the ROI was set as large as possible (3 × 3 mm). Orbital fat elasticity was used as an internal control. Subsequently, the elastic ratio was calculated by dividing the value obtained from the extraocular muscles by that of the orbital fat. A higher elastic ratio indicates greater stiffness in the extraocular muscles, which correlates with more advanced stages of fibrosis. Two investigators performed the extraocular muscles evaluation with real-time tissue elastography. We evaluated the images only when the pressure stabilized within the optimal 5-7 bar range.

Statistical analysis Statistical analyses were conducted using SPSS version 23 (IBM, Chicago, Illinois). Continuous variables are presented as mean ± standard deviation (SD), while categorical variables are reported as frequencies and percentages. Baseline characteristics, native T1 values, and elastic ratios were compared using analysis of variance (ANOVA) for continuous variables or chi-square tests for categorical variables, as appropriate. Our study selected the medial and lateral rectus muscles for data analysis, because of their ease in sonoelastographic imaging. The association between native T1 and elastic ratio was assessed using Pearson correlation analysis. Intraobserver agreement of MR parameters and elastic ratio were assessed using Bland-Altman methods and intraclass correlation coefficients. Receiver-operating characteristic (ROC) curves were built to determine the best threshold for quantitative elastic ratio to detect extraocular muscle fibrosis. ECV cutoff value of 48% was used as standard reference to determine the best threshold for elastic ratio value to detect extraocular muscle fibrosis in inactive TED. The optimal cutoff point was identified using the Youden index. All statistical tests were two-tailed, and a P value of <0.05 was considered statistically significant.