Association between individual retinal layer thickness and visual acuity in patients with epiretinal membrane: a pilot study

Abstract

Purpose

We investigated the correlation between visual acuity (VA) and individual retinal layer thickness in the foveal, parafoveal, and perifoveal regions of patients with an idiopathic epiretinal membrane (ERM).

Methods

One hundred and five subjects presenting with unilateral idiopathic ERM were included in this study. We segmented each patient’s optical coherence tomography (OCT) image into seven layers and calculated the mean layer thickness in the foveal, parafoveal, and perifoveal regions using the Iowa Reference Algorithm. In 105 patients with ERM, we detected correlations between their macular regions’ individual retinal layer thickness and their best corrected VA. Thirty-one of the 105 patients with ERM underwent vitrectomy and completed six months of follow-up. We then compared the 31 surgical patients’ preoperative and postoperative individual retinal layer thickness in each macular region. Additionally, the association between preoperative individual retinal layer thickness in each macular region and VA six months post-surgery in patients with ≥ two Snellen lines of visual improvement was determined.

Results

Multiple linear regression analysis showed that the inner nuclear layer (INL) thickness in the foveal, parafoveal, and perifoveal region were all associated with VA in the 105 patients ( R ² = 0.344, P < 0.001; R ² = 0.427, P < 0.001; and R ² = 0.340, P < 0.001, respectively). Thirty-one surgical patients 6 months post-surgery showed significantly decreased thicknesses ( P ≤ 0.012) of the foveal INL, inner plexiform layer (IPL), and outer nuclear layer (ONL); the parafoveal retina nerve fiber layer (RNFL), IPL, INL, and ONL; and the perifoveal RNFL, IPL, INL, ganglion cell layer (GCL), outer plexiform layer (OPL), and photoreceptor layer (PRL). We found a weak correlation between postoperative VA and preoperative foveal and perifoveal RNFL thickness ( r = 0.404 and r = 0.359, respectively), and a moderate correlation between postoperative VA and preoperative foveal and parafoveal INL thickness ( r = 0.529 and r = 0.583, respectively) in the 31 surgical patients ( P ≤ 0.047). The preoperative INL thickness in the foveal, parafoveal, and perifoveal regions showed a moderate to strong correlation ( r = 0.507, 0.644, and 0.548, respectively), with postoperative VA in patients with ≥ 2 lines of visual improvement ( P ≤ 0.038).

Conclusion

We detected a correlation between retinal damage and VA in the parafoveal, perifoveal, and foveal regions. Our results suggest that INL thickness in all macular regions may be a prognostic factor for postoperative VA in ERM patients.