Outcomes after Failed Pneumatic Retinopexy for Retinal Detachment.

PURPOSE: To provide visual and anatomic outcomes for patients with retinal detachment (RD) in whom primary pneumatic retinopexy (PR) failed. DESIGN: Retrospective, single-center, consecutive case series. PARTICIPANTS: Eyes with RD that failed a primary PR. METHODS: Anatomic and functional outcomes were evaluated for patients receiving treatment for failed PR. Three secondary procedures were compared, including repeat PR, pars plana vitrectomy (PPV), and combined scleral buckle (SB) plus PPV (SB+PPV). MAIN OUTCOME MEASURES: Anatomic reattachment and visual acuity (VA) at 1 year. RESULTS: Of a total of 423 primary PRs performed for RD, this study included 73 cases that failed. The overall single surgery anatomic success rate for the secondary procedure was 75%; the final success rate at 1 year was 100%. There was no statistically significant difference in success rates between repeat PR (63%), PPV (76%), and SB+PPV (88%). Improvement in VA was similar at 1 year between all 3 groups. Visual acuity at 1 year was similar between eyes undergoing PPV and SB+PPV (0.47 logMAR VA [Snellen equivalent, 20/59] for PPV and 0.52 logMAR VA [Snellen equivalent, 20/66] for SB+PPV; P = 0.75). Visual acuity at 1 year was better for those without macular involvement at the time of secondary procedure compared with eyes whose maculae detached (0.29 logMAR VA [Snellen equivalent, 20/39] vs. 0.73 logMAR VA [Snellen equivalent, 20/106]; P < 0.005). Fifty percent of PR failures underwent a secondary procedure within 1 week of primary PR; 80% occurred within 1 month. CONCLUSIONS: Anatomic success rates for secondary PR, PPV, and SB+PPV after failed PR were lower than published success rates for their use in primary RD. This suggests that a failed primary PR selects for RDs that are inherently more difficult to reattach. There was a trend suggesting that anatomic success rates are greater with SB+PPV than PPV and, in turn, with PPV than repeat PR. However, these differences were not statistically significant and did not translate into better VA gains at 1 year for either procedure. The suitable procedure after failed PR thus depends on patient presentation, surgeon preference, and patient preference.