Japanese journal of ophthalmology. 2023;67(6):699-710.
PURPOSE: To assess the efficacy and safety of periocular injections of methotrexate versus triamcinolone in the management of active thyroid-associated orbitopathy.
STUDY DESIGN: Prospective, double-masked, randomized clinical trial.
METHODS: Participants with bilateral active, moderate-to-severe thyroid-associated orbitopathy were randomly assigned to receive three periocular injections of 7.5 mg methotrexate in one orbit and three periocular injections of 20 mg triamcinolone in the contralateral orbit.
RESULTS: Among the enrolled 25 patients, 18 patients completed the study. A statistically significant reduction of the mean clinical activity score was detected in both arms (from 5.2 ± 0.89 at baseline to 0.9 ± 1.7 at study endpoint, p-value < 0.001 in the methotrexate arm, and from 5.1 ± 0.9 at baseline to 1 ± 1.7 at study endpoint, p-value < 0.001 in the triamcinolone arm), mean proptosis also decreased in both arms (from 25.2 ± 3.4 mm at baseline to 23.8 ± 3.7 mm at study endpoint, p-value = 0.01 in the methotrexate arm, and from 24.2 ± 3.06 mm at baseline to 23.2 ± 3.3 mm at study endpoint, p-value = 0.049 in the triamcinolone arm). Lid aperture and soft tissue signs improved significantly in both arms in comparison to baseline. A statistically significant reduction in the intraocular pressure was observed in the methotrexate arm but not in the triamcinolone arm. 88.9% of patients in both arms were overall responders at 6 months. There was no significant difference in mean CAS, proptosis, lid aperture or rate of responders between the two arms at any visit. Both drugs were found to be safe with minimal local and systemic complications.
CONCLUSION: Periocular injections of methotrexate represent an effective and safe alternative option for the management of active, moderate-to-severe thyroid-associated orbitopathy. Although no serious complications occurred during the 6-month follow-up, the possibility of late complications such as orbital fat atrophy cannot be ruled out.