Epiretinal Membrane Surgery: Structural Retinal Changes Correlate with the Improvement of Visual Function

Abstract

Several parameters have been studied for identifying the visual outcomes after pars plana vitrectomy (PPV) for epiretinal membrane (ERM) peeling. This retrospective study aimed to analyze structural retinal changes with spectral domain-optical coherence tomography (SD-OCT) and their correlations with visual acuity improvement in patients with ERM undergoing PPV. Twenty-one pseudophakic eyes were enrolled in the study. Ophthalmic evaluations included best corrected visual acuity (BCVA) and retinal layer thickness measurements with SD-OCT. The segmentation of the retina was divided into four parts: the inner retinal layer (IRL), composed of an internal limiting membrane, retinal nerve fiber layer, ganglion cell layer, and inner plexiform layer; inner nuclear layer (INL); outer plexiform layer (OPL); and outer nuclear layer (ONL). Correlations between changes in retinal layer thicknesses and BCVA were explored over a 6 month follow-up period. The postoperative thickness decrease of the INL was significantly correlated with thickness changes in all other measured retinal layers (p < 0.001). Non-parametric linear regression showed that postoperative improvement in BCVA was associated with a postoperatively decreased thickness in the IRL (p = 0.021), INL (p = 0.039), and OPL (p = 0.021). In eyes undergoing PPV, postoperative thickness decreases of the IRL, INL, and OPL were correlated with visual acuity improvements. Re-compaction of these retinal layers after relieving ERM-induced traction may be an important factor in postoperative visual function improvement.

Keywords: automated segmentation; optical coherence tomography; pars plana vitrectomy; single retinal layer; visual acuity.

Figure 1

Spectral domain optical coherence tomography scan showing the epiretinal membrane traction causing foveal distortion and alterations of the inner and outer retinal structure of the four retinal layers (the layer between the red and blue lines corresponds to the inner retinal layer; the layer between the blue and yellow lines corresponds to the inner retinal layer; the layer between the yellow and salmon lines corresponds to the outer retinal layer; and the layer between the salmon and pink lines corresponds to the outer nuclear layer) obtained using automated segmentation algorithm preoperatively (a) and postoperatively (b). In this latter scan (b), there is the presence of a reduction in retinal layer thickness and the foveal contour is partially restored. (ILM, inner limiting membrane; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer).

Figure 2

Mean and standard error of best corrected visual acuity (BCVA) during follow-up. The p-value reported in the figure is relative to the non-parametric Friedman test for repeated measures; * denotes a p-value < 0.05, *** denotes a p-value < 0.001 pairwise post-hoc analysis vs. previous measurement.

Figure 3

Correlation of average value of different layers of analyzed patients. Correlations were graphically depicted as correlation matrixes, using Tau correlation coefficients. Statistical significance was set at p < 0.05.

Figure 4

Non-parametric linear regression between variation of best corrected visual acuity and relative variation of layers thickness (expressed as average of different sector).