Controversy of indomethacin eye drops in the treatment of rheumatoid arthritis-induced corneal ulceration: a case report.

Background: Perforation of the cornea is a rare finding in patients with rheumatoid arthritis (RA). Addressing a perforated cornea associated with RA is challenging, since its pathogenesis is not fully elucidated. Topical nonsteroidal anti-inflammatory drugs (NSAIDs) were developed to prevent cystoid macular edema following cataract surgery in patients at risk. Their prescription in inflammation of the anterior segment of the eye may induce negative effects on the ocular surface. We bring into focus a corneal perforation in a patient with RA who used indomethacin eye drops to treat corneal ulceration, but responded promptly to drug discontinuation and initiation of topical cyclosporine 0.1%. Our aim is to emphasize two issues: the contraindication of topical indomethacin in corneal defects, and the immediate positive impact of topical cyclosporine 0.1% on corneal healing.

Case Presentation: A 73-year-old Caucasian woman with a 13-year history of RA was treated for corneal ulceration in her oculus sinister (OS) with topical indomethacin and gentamicin. The patient was being treated with systemic immunosuppression and NSAIDs for the underlying RA and artificial tears in both eyes. No bandage contact lens was used. After 3 weeks of treatment, perforation of the left cornea occurred and the patient was referred to our hospital. Upon admission, visual acuity (VA) in the OS was 20/630. Slit lamp examination of the OS revealed paracentral corneal perforation, iris plugging the perforation site, shallow anterior chamber, clear aqueous humor, and clear lens. Anterior segment optical coherence tomography showed the inclavated iris in the perforation site and minimum corneal thickness of 101 µm. Topical NSAIDs were discontinued and topical treatment was initiated with tobramycin, tropicamide 1%, phenylephrine 10%, and artificial tears five times a day, and occlusive patch. For 5 days, there was no improvement, so topical cyclosporine 0.1% was started, one drop every evening. Within 7 days, the cornea had healed, the iris was liberated from the perforation site, the minimum corneal thickness increased to 250 µm, VA improved to 20/25, and the patient was free of symptoms.

Conclusions: The main "takeaway" lessons from this case are that topical indomethacin should not be prescribed in cases of inflammation of the anterior segment of the eye, and that topical cyclosporine was efficacious in healing corneal perforation in our patient.