Clinical Study Hyperreflective Intraretinal Spots in ...

1 Hindawi Publishing CorporationJournal of diabetes ResearchVolume 2013, Article ID 491835,5 StudyHyperreflective Intraretinal Spots in Diabetics withoutand with Nonproliferative diabetic Retinopathy: AnIn VivoStudy Using Spectral Domain OCTS tela Vujosevic,1 Silvia Bini,1 Giulia Midena,2 Marianna Berton,1 Elisabetta Pilotto,1and Edoardo Midena1,31 Department of Ophthalmology, University of Padova, Via Giustiniani 2, 35128 Padova, Italy2 UniversityCampusBiomedico,ViaAlvarodelPo rtillo21,00128 Roma, ,ViaLivenza3,00198 Roma,ItalyCorrespondence should be addressed to Edoardo Midena; 20 August 2013; Revised 11 November 2013; Accepted 13 November 2013 Academic Editor: Ahmed M.

2 Abu El-AsrarCopyright 2013 Stela Vujosevic et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly To evaluate the presence of Hyperreflective Spots (HRS) in diabetic patients without clinically detectable retinopathy (noDR) or with nonproliferative mild to moderate retinopathy (DR) without macular edema, and compare the results to 36 subjects were enrolled: 12 with no DR, 12 with DR, and 12 normal subjects who served as controls.

3 All studied subjectsunderwent full ophthalmologic examination and spectral domain optical coherence tomography (SD-OCT). SD-OCT images wereanalyzed to measure and localize HRS. Each image was analyzed by two independent, masked The number ofHRS was significantly higher in both diabetics without and with retinopathy versus controls ( < ) and in diabetics withretinopathy versus diabetics without retinopathy ( < ). The HRS were mainly located in the inner retina layers (inner limitingmembrane, ganglion cell layer, and inner nuclear layer).

4 The intraobserver and interobserver agreement was almost perfect ( > ).Conclusions. SD-OCT Hyperreflective Spots are present in diabetic eyes even when Clinical retinopathy is undetectable. Theirnumber increases with progressing retinopathy. Initially, HRS are mainly located in the inner retina, where the resident microgliais present. With progressing retinopathy, HRS reach the outer retinal layer. HRS may represent a surrogate of microglial activationin diabetic IntroductionAn increasing body of evidence suggests that retinal neurode-generation and inflammation occur in human diabetes evenbefore the development of Clinical signs of diabetic retinopa-thy (DR) [1].

5 Retinal neural cell loss (neurodegeneration) hasalready been demonstratedin vivo(as thinning of retinalnerve fiber and ganglion cell layers), both in type 1 and 2diabetes [2 7].Retinal microglia activation has been recognized as themain responsible for the initial inflammatory response, eventhough the exact mechanism through which inflammatorycytokines are released remains poorly known [8]. Someexperimental studies have shown that retinal inflammationoccurring during the course of diabetes mellitus is a relativelyearlyeventandthatitprecedesbot hvasculardysfunctionand neuronal degeneration [1,8].

6 Joussen at al. demonstratedin animal models of diabetes mellitus that ICAM-1- andCD18-mediated leukocyte adhesion is increased in the retinalvasculature and accounts for many of the signature lesionsof DR [1]. Ibrahim et al. demonstrated in rats that theaccumulation of Amadori-glycated albumin (AGA) withinthe 8-week diabetic retina elicits microglial activation andsecretion of Tumor necrosis factor alpha (TNF- )[8].Retinal macroglia and retinal microglia activation havebeen documented histopathologically and hypothesizedinvivo[7,9 11]. The activated microglia secretes cytokines andother proinflammatory molecules used for the phagocytosisand the destruction of damaged cells as well as for thetriggering of reparative processes which lead to the formation2 Journal of diabetes Researchof glial scars [8].

7 If microglia remains in an activated state,continuously released cytokines may damage the neigh-bouring cells particularly the neuronal and the vascularones, leading to the onset of different retinal changes [8].According to this hypothesis, some histopathological studies(performed both in animals and in humans) have confirmedthe activation of microglial cells, as well as the presenceof different inflammatory molecules secreted by microglia,commonly associated with neuronal and endothelial death[9,10,12 14].Spectral domain optical coherence tomography (SD-OCT) has become a valuable tool for thein vivoevaluationof single retinal layers (both the inner retina and the outerretina) in diabetic patients [7,15,16].

8 Moreover, it has beenused for the evaluation of Hyperreflective retinal Spots in agerelated macular degeneration, diabetic macular edema, andretinal vein occlusion [16 21].The main purpose of this Study was to determine,invivo, by SD-OCT, the presence and location of hyperreflectivespots in the retina in diabetic patients without DR or withearly stages of DR (mild and moderate nonproliferative DR)without macular edema versus normal Material and Methods36 subjects were included in the Study : 12 subjects servedas controls, 12 patients were affected by diabetes withoutdiabetic retinopathy (no DR), and 12 patients were affectedby diabetes and diabetic retinopathy (mild to moderate).

9 Oneeye of each subject was used for the SD-OCT analysis. Theexclusion criteria were proliferative DR, macular edema, anytype of previous retinal treatment (macular laser photoco-agulation, vitrectomy, intravitreal steroids, and/or antiangio-genic drugs), any intraocular surgery, refractive error>6D,previous diagnosis of glaucoma, ocular hypertension, uveitisor other retinal diseases, and significant media opacitiesthat precluded fundus imaging. All patients underwent SD-OCT using Spectralis (Heidelberg Engineering, Heidelberg,Germany). A single 180 SD-OCT line scan (6 mm length)centered onto the fovea was analyzed for each patient, lookingfor the presence of Hyperreflective Spots .

10 Two red verticallines were traced at 500 mand1500 mfromthecenterofthe fovea in the temporal region, thus excluding the fovealavascular zone. A manual count of the Hyperreflective Spots ,defined as small, punctiform, white lesions, was performedbetween the two markers. The layering was obtained usingthe automatic layering of the Spectralis SD-OCT with manualrefinement for the boundaries of the most critical layers ( ,inferior boundary of ganglion cell layer where contrast islower).The count was performed starting from the inner lim-iting membrane (ILM) to the outer nuclear layer (ONL),including ILM to ganglion cell layer (GCL); inner nuclearlayer (INL) to outer plexiform layer (OPL), and ONL.