Concordância entre a angiografia por OCT e a avaliação multimodal na detecção de neovascularização de coroide no descolamento do epitélio pigmentado da retina

• Autor(a) principal: Oliveira, Tauan de
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFG
• Texto Completo: http://repositorio.bc.ufg.br/tede/handle/tede/9813
• topic: Descolamento do epitélio; Imagem multimodal; Angiografia por tomografia de coerência óptica; Pigmentado da retina; Neovascularização de coroide; Angiografia com indocianina verde; Angiografia com fluoresceína; Optical coherence tomography angiography; Multimodal imaging; Retinal pigment epithelium detachment; Choroidal neovascularization; Indocyanine green angiography; Fluorescein angiography; CLINICA MEDICA::OFTALMOLOGIA

The aim of this study was to evaluate the agreement between multimodal imaging – MI (fluorescein angiography, indocyanine green angiography, optical coherence tomography - OCT) and optical coherence tomography angiography (OCTA) in the detection of choroidal neovascularization (CNV) in patients with pigment epithelial detachment with subretinal/intraretinal fluid (PED+F) and patients with PED without subretinal/intraretinal fluid (PED-F). We evaluated 22 eyes of 15 patients were divided into 2 groups (PED+F and PED-F). All patients underwent MI and OCTA with manual and automatic segmentations. OCTA findings were conferred to MI findings and then analyzed for diagnostic concordance. In the PED+F group (10 eyes) all studied eyes demonstrated CNV in MI. In manual segmentation OCTA assessment, 9 of 10 eyes (90%) were detected with CNV. When evaluated by automatic segmentation, 8 of 10 eyes (80%) revealed the presence of CNV. In the PED-F (12 eyes) group, all eyes did not demonstrate CNV in MI and OCTA evaluations, either by manual or automatic segmentation. The agreement between MI and OCTA has shown concordance (k: 0.908; 95% CI, 0.491-1.000); the evaluation of the agreement between the automatic and manual segmentation also showed concordance (k: 0.904; 95% CI, 0.488-1.000). The solid agreement between the multimodal imaging regarding the ability of OCTA to identify possible initial CNV in a patient with PED-F was observed. OCTA accuracy in the detection of CNV was 95.45%. In addition, the solid agreement between manual and automatic segmentations to identify CNV on OCTA was also shown.