The incidence of corneal abnormalities in the Silicone Study. Silicone Study Report 7.

OBJECTIVE

To determine factors that were prognostic of corneal abnormalities in eyes following surgery for severe proliferative vitreoretinopathy.

DESIGN

Subgroup analysis of the Silicone Study.

SETTINGS: Community and university-based ophthalmology clinics.

MATERIALS: Eyes with attached maculae at the 24-month follow-up examination that did not have a pre-existing corneal abnormality.

INTERVENTIONS: Vitrectomy surgery with long-acting gas or silicone oil.

OUTCOME MEASURES: Epithelial and/or stromal edema, corneal opacity and/or previous corneal transplant.

RESULTS

The incidence of corneal abnormalities at 24 months was 27% and did not differ significantly between treatment groups. Prognostic factors were preoperative aphakia or pseudophakia (P = .003), preoperative iris neovascularization (P = .006), reoperation (P = .001), the absence of a fluid/gas exchange (P = .03), corneal touch by silicone oil (P = .02), and the presence of aqueous cells (P = .009) or aqueous flare (P = .08). In a multivariate analysis, independent prognostic factors were iris neovascularization (relative risk [RR] = 13.1), aphakia or pseudophakia (RR = 3.0), postoperative aqueous flare (RR = 5.4), and reoperations (RR = 3.4). Corneal abnormalities were correlated with poor visual acuity and hypotony (P < .001).

CONCLUSIONS

To our knowledge, this is the first study to document that the incidence rates of corneal abnormalities are equivalent between oil and gas. The incidence of corneal abnormalities in gas-filled eyes was higher than expected, and remained high in oil-filled eyes, despite the use of an inferior iridectomy. Successful surgical repair of the retinal detachment with a single operation, and prevention and early management of corneal touch by silicone oil should help to prevent corneal abnormalities. If rubeosis iridis or severe aqueous flare is present, preoperative treatment with intense topical and possibly periocular steroids might reduce inflammation, which might mediate corneal damage.