Re: Zhao et al.: Pars plana vitrectomy and endoresection of refractory intraocular retinoblastoma ( Ophthalmology . 2018;125:320-322)

Abstract

questions, which we decided to perform morphologic immunocytochemical studies, marking each cell and cellular contents of the outer retinal layers with specific antibodies. CRALBP (cellular retinaldehyde-binding protein) stains the whole retinal pigment epithelium (RPE) cell body and processes, allowing clear visualization of the apical processes and interdigitations ensheathing the cone outer segments. In our preparations, we do not see any structure similar to the contact cylinder described by Spaide and Curcio. 3 We center our study on the fovea, where interdigitations of the RPE extend the entire length of cones outer segments and reach the cone ellipsoid level (Fig 4A, C, D, and Fig 5 in the original article). The electron microscopy study to which Curcio et al refer in their letter shows structures in the perifoveal areas, where cones outer segments are shorter and their tips do not reach the RPE. Also, as they state, manipulation and the integrity of cones and RPE processes during the preparation of transmission electron microscopy samples could create certain artifacts difficult to interpret.So, if interdigitations in the fovea extend and reach the ellipsoids, which are thought to be the second band, the third band cannot be the interdigitation zone because there would be no anatomic structure in between that could explain the hyporeflectivity. Then, we discard the interdigitation zone as responsible for band 3. Instead, we propose that phagosomes can explain the third band hyperreflectivity. As seen in Fig 6D-F in the original article, they are located in the RPE apical zone, within the cell bodies (Fig 6E, F). The possibility that phagosomes are what produce the band 3 hyperreflectance has not been described to date and could explain some changes of this band that were difficult to explain with previous interpretations. 5 In our model, band 4 hyperreflectivity is explained by the basal RPE, where the mitochondria accumulate, together with the Bruch's membrane. We believe the space between phagosomes (apical RPE, band 3) and mitochondria (basal RPE, band 4) correspond with the hyporeflective band. In this space, we find (apart from melanosomes, lipofuscin, and melanolipofuscin) the nucleus, cytoplasm, and other RPE cell organelles. In fact, all the inner retinal hyporeflective OCT bands correspond with nuclear layers, the same as we are proposing here for the melanosome zone, where we include the RPE nucleus.Of course, we agree with Curcio et al that more experimental data must be obtained to verify our proposal, but we believe we cannot close the discussion yet. We should consider this as an open field for new interpretations, taking into account that the interpretation that Curcio et al propose has still some unclear aspects that should be clarified. We are sure that next technologies including adaptive optics OCT would help to clarify the interpretation.