Comparison of Long-Term Automated Retinal Layer Segmentation Analysis of the Macula between Silicone Oil and Gas Tamponade after Vitrectomy for Rhegmatogenous Retinal Detachment

Abstract

Purpose: To identify long-term changes in individual retinal layer thickness using automated retinal layer segmentation analysis on high-resolution spectral-domain optical coherence tomography (SD-OCT) scans of eyes with macula-off rhegmatogenous retinal detachment (RRD) treated with vitreoretinal surgery (VRS) and gas or silicone oil tamponade and having single-operation success. Methods: A total of 58 patients operated on by VRS for RRD and followed up for 12 months were imaged by SD-OCT. The patients with retinal diseases such as an epiretinal membrane or cystic macular edema in the operated and fellow eyes were excluded. The thicknesses of the retinal nerve fiber layer (RNFL), ganglion cell layer (GCL), inner plexiform layer (IPL), inner nuclear layer (INL), outer plexiform layer (OPL), outer nuclear layer (ONL), photoreceptor layer, and retinal pigment epithelium were compared to those of the fellow eyes after the 12-month follow-up. Thickness changes in individual layers were quantitatively analyzed in the operated and fellow eyes and correlated with the type of tamponade used in the surgery. Results: Spectralis OCT automated segmentation software was used for the retinal layer analysis. There were 22 females and 36 males. Their mean age was 60.7 ± 11.2 years. The mean central macular thickness was 214.3 ± 29.5 µm in the operated and 229.7 ± 21.7 µm in the fellow eyes (p = 0.008). There was a statistically significant difference between the operated and the healthy fellow eyes in the following layers: the RNFL (p = 0.017), GCL (p = 0.02), INL (p = 0.005), and ONL (p = 0.008) in the central foveal area; the RNFL (p < 0.001), INL (p = 0.017), and ONL (p = 0.022) in the perifoveal ring; and the RNFL (p < 0.001), IPL (p = 0.042), INL (p = 0.001), and OPL (p = 0.001) in the peripheral ring. The logMAR best corrected visual acuities were 2.51 ± 0.68 and 2.69 ± 0.62 at baseline and 0.60 ± 0.38 and 0.50 ± 0.38 at month 12 in the silicone oil tamponade (n = 28) and the gas tamponade (n = 30) group (p = 0.52 and p = 0.21, respectively). The foveal GCL, OPL, and ONL and the perifoveal GCL and IPL were statistically significantly thinner in the silicone oil tamponade group (p = 0.01, p = 0.046, p = 0.024, p = 0.006, and p = 0.011, respectively). Conclusions: Significant changes were observed in the retinal layers after VRS for RRD. Individual retinal layers seem to be affected 1 year after VRS for RRD. The type of tamponade can influence the thickness of the retinal layers. The thickness of the retinal layers was significantly preserved in eyes treated with gas tamponade when compared to those treated with silicone oil tamponade in the long term. Further studies are needed to validate our results.