Vitreoschisis in Diabetic Macular Edema

A recent publication on OCT imaging of the vitreoretinal interface in 44 patients with diabetic macular edema (DME) described that half had anomalous posterior vitreous detachment (PVD), with membranes attached to the macula and/or optic disc. There were several images, especially Figure 3A, which the authors interpreted as a “relatively thick EVi membrane, which is overlying cystoid macular edema and tiny amount of subretinal fluid, splits and continuous with both a detached hyperreflective membrane at the posterior vitreous cortex and a broken and thinner hyperreflective adherent ERM.” Without clinicopathologic correlation, this interpretation is mere conjecture. In Figure 2, the authors claim to see “vascularization ”; however, the OCT image does not demonstrate blood vessels, and no histopathology is presented. Regarding Figure 4, the authors state, “The B-mode SD-OCT discloses a full-thickness split in the relatively thin unified ERM/PViC membrane coupled with its retraction and rolling. ” This is a contradiction in terms, as there cannot be “full-thickness” tissue after it splits, a feature that would have been revealed by histopathologic analysis, had it been performed. Indeed, although the clinical observations of this study may be valid, the interpretation of the findings is flawed due to a lack of clinicopathologic correlation and a failure to appreciate the role of diabetic vitreopathy in advanced disease. Figure 1 demonstrates a spectral OCT-SLO image obtained from a patient with macular pucker. The vitreoretinal interface in this patient is nearly identical with that in Figure 3A of the aforementioned study, with a clearly visible split in the posterior vitreous cortex, known as “vitreoschisis.” In contrast to the aforementioned study, however, this patient underwent clinicopathologic correlation, and the surgically excised tissue was analyzed histopathologically. The results confirmed schisis of the posterior vitreous cortex. Vitreoschisis was first described in proliferative diabetic retinopathy (PDR) by ultrasound and by histopathology, where 80% of eyes had a split in the posterior vitreous cortex. Spectral OCT-SLO studies have furthermore shown that half of eyes with macular hole and macular pucker have evidence of vitreoschisis. In the aforementioned study of DME, 13 (57%) of the 23 subjects with anomalous PVD had PDR. This is an important consideration, given the high prevalence of vitreoschisis in this group of patients. Thus, the findings at the vitreoretinal interface in the aforementioned study of DME are most likely the manifestation of diabetic vitreopathy with vitreoschisis. Indeed, Figure 3A in their publication is an excellent example of vitreoschisis. The one noteworthy finding in this study is the detection of vitreopapillary adhesion (VPA) in 19 (83%) of the 23 eyes with anomalous PVD and DME. Previous studies have shown a similarly high prevalence of VPA in macular holes, lamellar holes with cysts, and macular pucker with cysts. As previously described, VPA influences the vector of force on the macula, resulting in outward (centripetal) tangential traction inducing macular holes and cystoid spaces in vitreomaculopathies, perhaps in DME as well. This is important, because it has long been presumed that the cysts in DME are fluid-filled manifestations of exudation. It is likely, however, that in cases with VPA there may be a tractional component inducing cystoid spaces that, as in macular holes (full-thickness and lamellar), are not due to exudation. With respect to the authors’ remaining interpretations, we must recognize and appreciate the difference between what is seen and what is reality. In The Allegory of the Cave, Socrates spoke of our sensory experiences as mere shadows on the wall in the cave of our misconceived world. Appreciating the limitations of our senses, even when enhanced by imaging technology, leads to the conclusion that our view of the world is just a model and as such is suscep-