Author Block: Bryon R. McKay .  University of Toronto.

Author Disclosure Block: B.R. McKay:  None.

Abstract Title: -Year Evaluation of Subretinal Fluid Drainage Techniques in Pars Plana Vitrectomy for Primary Rhegmatogenous RetinalDetachment (ELLIPSOID-2 Study)

Abstract Body: Purpose:  To compare visual acuity, complications, and photoreceptor integrity following pars plana vitrectomy (PPV) with subretinal fluid (SRF) drainage from the peripheral retinal breaks (PRB) vs. posterior retinotomy (PR) vs. perfluorocarbon liquid (PFCL) for rhegmatogenous retinal detachment (RRD) at 2 years post-surgery   Study Design:  Retrospective analysis of 300 eyes (300 patients) with primary uncomplicated RRD that underwent PPV (100 consecutive patients included in each group), 2-year extension from the ELLIPSOID study.   Methods:  Primary outcomes were visual acuity (BCVA) and complications (cystoid macular edema (CME), epiretinal membrane (ERM)). Secondary outcomes were discontinuity of the external limiting membrane (ELM), ellipsoid zone (EZ) and interdigitation zone (IDZ) on spectral-domain optical coherence tomography (SD-OCT) at 2-year post-surgery   Results:  Proportion of patients with visual acuity assessment and gradable SD-OCT at 2-years was similar between group (PRB 91%, PR 86%, PFCL 87%). There were no significant differences in age, sex, baseline visual acuity, lens status, extent of RRD and time from macula-off to presentation. Single-operation reattachment rate at 24 months was similar between groups (PRB 86%; PR 85%; PFCL 83%, p=0.9). Mean(±SD) logMAR BCVA at 24 months was significantly better in the PRB compared to PR and PFCL, with PFCL having the worst BCVA (PRB 0.5±0.6; PR 0.7±0.5; PFCL 0.9±0.7, p=0.001). Rate of CME was significantly higher with PFCL (PRB 29.7%; PR 30.2%; PFCL 45.9%, p=0.0015) and ERM formation was significantly higher in PR (PRB 62.6%; PR 93.0%; PFCL 68.9%, p=0.002) compared to the other groups. There were no differences in ELM or EZ discontinuity on the 3-mm foveal scans. However, discontinuity of the IDZ was significantly higher in the PFCL group (PRB 34%; PR 27%; PFCL 46%, p=0.002) on the 3-mm foveal scans at 24 months   Conclusions:  Visual acuity at 2-years was inferior in eyes with PFCL compared with drainage from the PRB or PR for primary uncomplicated RRD. Drainage via PRB demonstrated the best visual acuity recovery. There was a corresponding greater risk of discontinuity of the IDZ in PFCL cases, and a greater risk of CME with PFCL and greater ERM with PR. Drainage technique may impact long-term visual acuity results and photoreceptor integrity