THU0560 DISTINCTIVE IMAGING FEATURES BETWEEN IGG4-RELATED OPHTHALMIC DISEASE AND GRAVES' ORBITOPATHY: A COMPARATIVE STUDY

Abstract

Background: IgG4-related ophthalmic disease (IgG4-ROD) may present as a cause of orbital myositis leading to proptosis and diplopia. This clinical scenario can be mistakenly diagnosed as Graves' orbitopathy (GO), preventing a timely and adequately treatment. Objective(s): To elucidate if there are specific radiological features that might differentiate between IgG4-ROD and GO by imaging. Method(s): We included 19 patients with diagnosis of IgG4-related disease (IgG4-RD) according to the Comprehensive Diagnostic Criteria for IgG4- RD, who regularly attended a tertiary referral center in Mexico City. All the patients had ophthalmic involvement and available computed tomography (CT) and/or magnetic resonance imaging (MRI) of the orbits. We also included 32 patients with GO with available CT and/or MRI of the orbits. Imaging studies were evaluated by a blinded head and neck radiologist for the following features: exophthalmos, extraocular muscles (EOM) size and morphology, lacrimal gland enlargement, orbital fat involvement, stretching of the optic nerve (ON), ON sheath thickening and orbital bone changes. Result(s): Both groups were similar in age (49.1+/-15.8 vs. 51.6+/-14.7, p=0.58) and gender (men 58.9% vs. 40.6%, p=0.23). In addition to ophthalmic involvement, 18 (94.7%) IgG4-RD patients had extra-ophthalmic involvement with a median number of organs involved of 7 (1-12), mainly submandibular glands (73.7%), lymph nodes (68.4%), parotid glands (63.2%) and pancreas (47.4%). Three patients were misdiagnosed as GO before IgG4-RD diagnosis. Graves' disease was the underlying thyroid disorder in 28 (87.5%) GO patients, Hashimoto's thyroiditis in 2 (6.3%), papillary thyroid carcinoma in one (3.1%) and one patient was euthyroid with positive thyroid stimulating immunoglobulin. The prevalence of exophaltmos (78.9% vs. 93.8%), bilateral involvement (78.9% vs. 87.8%) and overall EOM involvement (47.4% vs. 68.8%) was similar between IgG4- ROD and GO groups. However, IgG4-ROD patients had a higher frequency of lacrimal gland involvement (73.7% vs. 10.7%, p=0.001) and a tendency for the lateral rectus to be the most frequently involved EOM (22.2% vs. 0%, p=0.07); conversely they had a lower prevalence for the inferior rectus to be the most frequently involved EOM, (33.3% vs. 72.7%, p=0.04), orbital fat involvement (47.4% vs. 81.3%, p=0.01), ON stretching (57.9% vs. 87.5%, p=0.02) and orbital bone changes (0% vs. 25%, p=0.02). EOM bellies were involved in all the IgG4-ROD and GO cases, whereas EOM tendon involvement was present in 9% of GO and in none of the IgG4-ROD group. Patients with IgG4-ROD had more frequently the combination of lacrimal gland and lateral rectus (31.6% vs. 3.1%, p=0.008) and less frequently the combination EOM and orbital fat involvement (21.2% vs. 59.4%, p=0.008). At the logistic regression analysis we found an association of lacrimal gland involvement (OR 64.4.0, 95% CI 6.8-609.5, p=0.001) with IgG4- ROD. In a second model including combined variables, the combination of lacrimal gland and lateral rectus involvement was associated with IgG4-ROD (OR 62.5, 95% CI 3.31-1000), P=0.006), whereas the presence of EOM and orbital fat involvement was protective (OR 0.05, 95% CI 0.006-0.48, p=0.009). Conclusion(s): Imaging features may reliably differentiate between IgG4-ROD and GO. The presence of both lacrimal gland and lateral rectus enlargement must alert clinicians to consider IgG4-ROD diagnosis.