Natural History of Retinal Vein Occlusions: Systematic Review and Meta-Analysis

Abstract

Purpose: : Retinal Vein Occlusion (RVO) is an important sight-threatening condition, however little data exists on the natural history of RVO. The purpose of this study was to conduct a review of the literature and meta-analysis to assess the natural history and progression of eyes with untreated RVO.

Methods: : A comprehensive, systematic review of all observational RVO natural history or interventional studies containing a control arm, published in English up to September 2007, was undertaken. We electronically searched MEDLINE (1950-2007), EMBASE (<1966-2007), Current Contents (1999-2007), the Cochrane Library and PubMED for the prior 6 months. The reference lists of accepted papers and reviews from the last 5 years were also reviewed. Weighted mean Visual Acuity (VA) at baseline (for studies with no VA selection criteria) and weighted mean change in VA were calculated using the Der Simonian and Laird random-effects model for meta-analysis. Summary estimates and 95% confidence intervals (95%CI) were calculated after 3, 6 and 12 months baseline. Snellen VA scores were converted to logMAR using standard calculations and logMAR values of 1.7, 2.0, 2.7 and 3.0 were substituted for all Snellen VAs of "Count Fingers", "Hand Movements", "Light Perception" and "No Light Perception", respectively.

Results: : Out of 5829 citations, 51 studies were included in the meta-analyses. Mean VA at baseline for branch RVO patients was 0.684 (95%CI 0.600 to 0.767) with no significant improvement from baseline after 3 (-0.56 lines, 95%CI -1.94 to +0.82), 6 (+ 0.35 lines, 95%CI -0.27 to +0.97) or 12 months (-1.54 lines, 95%CI -2.71 to +0.37). In patients with non-ischemic central RVO, mean VA at baseline was 0.492 (95%CI 0.356 to 0.627). VA improved significantly after 3 months (+2.15 lines, 95%CI +1.20 to +3.11) but was not significantly different after 6 (+0.51 lines, 95% -2.30 to +3.32) or 12 months (+1.58 lines, 95%CI -0.45 to +3.61). For ischemic central RVO patients, mean baseline VA was 1.578 (95%CI 1.351 to 1.804). Vision significantly improved after 6 months (+3.92 lines, 95%CI +1.77 to +6.07) but was not significantly different after 3 (+1.62 lines, 95% CI -2.06 to +5.3) or 12 months (+1.29 lines, 95% -0.51 to +3.09).

Conclusions: : The results from this meta-analysis show that vision of untreated RVO was poor at the time of presentation and had little improvement over 12 months.

Keywords: clinical (human) or epidemiologic studies: natural history • retina • visual acuity