AIM: To report the long-term outcomes of a large cohort of upper lid retraction patients treated with subconjunctival triamcinolone acetonide injections. METHODS: This retrospective clinical study was conducted from 2009 to 2016. The documents of 97 patients (126 eyes) were reviewed. The patients were treated with subconjunctival triamcinolone acetonide injections monthly and evaluated pre- and post-treatment by taking measurements of the upper margin reflex distance (MRD1, the distance between the upper-lid margin and the pupil center), lid aperture, lagophthalmos and proptosis. The patients were divided into the cured group, the improved group and the ineffective group based on the final treatment effectiveness. The improved group and the ineffective group were combined into non-cured group. RESULTS: The follow-up time was 23.45±14.86mo. Of the patients, 63 (64.9%) were assigned to the cured group, 22 (22.7%) in the improved group, 12 (12.4%) in the ineffective group. Injection times for cured group was 4.66±1.74, and for non-cured group was 5.17±2.08 (P≤0.05). The MRD1 was significantly reduced between baseline and 1mo after the initiation of treatment in the cured and improved groups (P<0.001). The values did not change significantly after 4mo for the improved group, while they continued to decrease in the cured group. In the ineffective group, the MRD1 did not significantly change after treatment. A statistically significant difference was observed between the baseline and post-treatment values of the mean muscle thickness (2.95±0.99 mm vs 2.04±0.63 mm, P<0.05) and surface area (27.27±10.5 mm2 vs 18.3±7.04 mm2, P<0.05) of the levator superioris with magnetic resonance imaging examination. Twelve patients (12.4%) relapsed after treatment and required additional subconjunctival injections. The main side effects included menstrual cycle disturbances (32%) and elevations of intra-ocular pressure (18.6%). CONCLUSION: A repeated subconjunctival injection of triamcinolone is an effective and relatively safe treatment for upper lid retraction due to thyroid-associated ophthalmopathy. This is related to the anti-inflammation effect of the medicine and the levator thickness can become thinner. Patients must be monitored regularly for recurrences and side effects.