Characterization and outcomes of repeat orbital decompression for thyroid-associated orbitopathy.

Orbital decompression for thyroid-associated orbitopathy (TAO) is commonly performed for disfiguring proptosis, congestion, and optic neuropathy. Although one decompression typically achieves goals, a small percentage requires repeat decompression. We performed a 10-year retrospective chart review of all orbital decompressions for TAO at a single tertiary referral institution. Four-hundred and ninety-five orbits (330 patients) were decompressed for TAO, with 45 orbits (37 patients) requiring repeat decompression. We reviewed the repeat cases for indications, clinical activity scores, approach, walls decompressed, and outcomes. Nine percent of orbits required repeat decompression for proptosis (70%), optic neuropathy (25%) or congestion (45%). Sixty-four percent were for recurrence of disease, 36% were for suboptimal decompression. Three incisional approaches were used: lateral upper eyelid crease, inferior transconjunctival, and transcaruncular, with inferior transconjunctival being most common. Of the three walls removed, deep lateral, inferior, and medial, the deep lateral wall was most common (51%). A repeat lateral decompression was the most frequent pattern. Of 37 patients requiring repeat decompression, 40% had diplopia prior to repeat, and an additional 24% developed diplopia after the repeat. Whereas previous studies published by our group cited only 2.6% of deep lateral wall orbital decompressions leading to new-onset primary gaze diplopia, repeat orbital decompressions have a much higher rate of post-operative diplopia. The new onset primary gaze diplopia after repeat decompression group had a higher average preoperative CAS (3.3 vs. 2.4, p < 0.01), higher mean blood loss (56 vs. 19 mL, p = 0.04), more frequent medial wall decompressions (47% vs. 29%, p = 0.33), and greater proptosis reduction (2.4 vs. 1.7 mm, p = 0.24).