Retinal endovascular surgery for central retinal vein occlusion: initial experience of four surgeons.

PURPOSE: The rate and magnitude of spontaneous visual recovery are very poor in central retinal vein occlusion (CRVO). In the first follow-up year, the Central Vein Occlusion Study Group reported that only 6% of eyes recovered > or = 3 lines of vision and none recovered > or = 8 lines of vision. Retinal endovascular surgery (REVS) is vitrectomy followed by cannulation of retinal vessels with injection of tissue plasminogen activator. After reports of one surgeon's experience suggested that the procedure promotes recovery of vision, other surgeons began to offer REVS to their patients. This report discusses the initial experience of four surgeons using REVS to treat CRVO. METHODS: In this prospective, consecutive case series, patients with CRVO for > 1 week and visual acuity of < 20/50 were offered REVS and were followed by the authors. The main outcome measure was recovery of visual acuity. RESULTS: This series represents the initial 25 consecutive REVS cases of the 4 authors (5-7 cases per author). The median CRVO duration was 2 months (mean 2.9, months), and the average preoperative visual acuity was 20/400 (< or = 20/200 in 80% of cases). Intravitreal triamcinolone acetonide (IVTA) was administered intraoperatively in three cases and at some time postoperatively in six cases. Overall, 18 eyes (72%) recovered > or = 3 lines of visual acuity, and 9 (36%) recovered > or = 8 lines of vision. Among the surgeons, the rates of > or = 3-line visual recovery ranged from 57% to 100%, and the rates of > or = 8-line visual recovery ranged from 14% to 71%. Of the 22 eyes that initially underwent REVS without intraoperative IVTA injection, 13 (59%) recovered > or = 3 lines of vision and 5 (23%) recovered > or = 8 lines of vision. Complications included macular edema (28%), anterior segment or retinal neovascularization (24%), and subsequent cataract surgery (5 [23%] of the 22 preoperatively phakic eyes). One eye had an intraoperative retinal detachment that was treated but recurred 4 months after REVS, and two of the eyes with anterior segment neovascularization developed late-onset traction retinal detachments (at 8 and 13 months after REVS). CONCLUSION: Although the authors were on the "learning curve" of experience during this series, REVS appears to promote visual recovery far in excess of what would be expected to occur spontaneously, and IVTA injection greatly improved outcomes. We believe that mastery of REVS techniques and the inclusion of IVTA injection may lead to better visual results and lower complication rates.