Abstract
Purpose:
To determine whether the morphologic appearance of the foveal contour and the continuity of the IS/OS junction as assessed with optical coherence tomography (OCT) of patients presenting with epiretinal membrane (ERM) correlate with changes in best corrected visual acuity (BCVA), central retinal thickness (CRT), central subfield thickness (CST) and volume (CSV) following pars plana vitrectomy (PPV) for ERM removal with internal limiting membrane (ILM) peel.
Methods:
Images from consecutive forty-seven subjects were analyzed retrospectively. Foveal contour grades were determined by 2 independent graders using radial line scans at 30° intervals. Grades were classified as 0 (fovea thinner than surrounding macula), 1 (flat), and 2 (fovea thicker than surrounding macula). In cases of observer disagreement, OCT-derived thickness values of the retina at the fovea and 1 mm radially from the fovea were used to generate quantifiable thickness profiles and applied as tiebreakers. The continuity of the IS/OS junction was assessed on 6 radial line scans. CST and CSV data were collected from OCT-generated map.
Results:
40 of the 47 subjects had both preoperative and 3 month follow-up OCT. 29 of these 40 were Grade 2 preoperatively, 11 were Grade 1, and none were Grade 0. 14 of the 29 Grade 2 subjects resolved to Grade 0/1, while 3 of the preoperative Grade 0/1 subjects were Grade 2 at 3 months (McNemar test, P=0.013).Preoperative CRT, CST, and CSV values were negatively correlated with the changes in CRT (P<0.001), CST (P<0.001), and CSV (P<0.001), respectively, at 90 days. Foveal contour grade was not significantly correlated to the change in these outcome values (P>0.05). BCVA was not found to improve significantly in preoperative Grade 0/1 patients (-0.11 LogMar units, P=0.122), nor in preoperative Grade 2 patients (-0.04, P=0.380), at 3 months. No significant difference in BCVA changes between groups was found (P=0.420). However, linear mixed model analysis with random subject effect demonstrated a statistically significant improvement in BCVA (regression coefficient = 0.01 LogMar units, P=0.048) when controlling for the confounding effects of lens status (0.18, P=0.001) and baseline IS/OS junction disruption (0.15, P=0.012).
Conclusions:
Preoperative IS/OS junction disruption and postoperative cataract progression, but not preoperative foveal contour grade, significantly affect BCVA following PPV for ERM removal with ILM peel.
Keywords: 585 macula/fovea •
550 imaging/image analysis: clinical •
762 vitreoretinal surgery