Hyperreflective pyramidal structures on optical coherence tomography in geographic atrophy areas.

PURPOSE: We observed hyperreflective dome-shaped or pyramidal structures (HPS) on spectral domain optical coherence tomography (SD-OCT) in patients affected with geographic atrophy (GA). Our purpose was to describe the multimodal imaging features of HPS identified in areas of GA in patients with age-related macular degeneration. METHODS: This is a retrospective case series of patients with GA harboring HPS in atrophic areas. Multimodal imaging examination including infrared reflectance, fundus autofluorescence, and SD-OCT, was performed for each patient. Infrared and fundus autofluorescence appearance and mean SD-OCT height of HPS in GA were analyzed. RESULTS: A total of 36 eyes of 25 patients (20 women; mean age, 82.3 ± 5.9 years, range, 73-92 years) with GA were included. A total of 96 HPS in GA were analyzed by SD-OCT. In all HPS (96/96, 100%), the peripheral part was hyperreflective. In 66 of 96 HPS (69%), the center was heterogeneously hyperreflective, whereas in 30 of 96 HPS (31%), the center was hyporeflective. On infrared reflectance images, HPS in GA appeared as hyporeflective lesions surrounded by hyperreflective halos, within an area of background hyperreflectivity because of GA in all eyes. On fundus autofluorescence, 39 of 96 HPS (41%) were heterogeneously hyperautofluorescent, whereas 57 of 96 HPS (59%) were hypoautofluorescent. Mean height of HPS was 91 ± 50.9 µm in the foveal scan (range, 42-291 µm). CONCLUSION: We describe a multimodal imaging of distinctive lesions that presented as hyperreflective pyramidal structures on SD-OCT. We suggest the name "ghost drusen" because these HPS appear in GA areas, and because of their pyramidal or dome-shaped aspect on SD-OCT.

Evaluation of semiautomated measurement of geographic atrophy in age-related macular degeneration by fundus autofluorescence in clinical setting.

PURPOSE: To assess intraobserver and interobserver agreement among physicians with different degrees of clinical experience, using a novel fundus autofluorescence semiautomated software for quantification of geographic atrophy in clinical setting. METHODS: Fundus autofluorescence frames (excitation: 488 nm; emission: 500-700 nm) of 29 eyes (20 patients; mean age, 79.6 ± 6.2 years) with geographic atrophy secondary to age-related macular degeneration, and no signs of choroidal neovascularization, were analyzed using Region Finder, a semiautomated software embedded in Spectralis (Heidelberg Engineering, Heidelberg, Germany). For each study eye, semiautomated atrophy identification and quantification were independently performed, twice (in a 2-week time frame), by 3 readers with different degrees of clinical experience (2 fellows, and 1 resident). Intraobserver and interobserver agreements were assessed. RESULTS: Mean difference in intraobserver agreement (Bland-Altman statistics) ranged from -0.17 mm² to 0.13 mm². Intraobserver agreement was excellent until the geographic atrophy threshold value of 15.72 mm². Variability correlated with the size of atrophy. Mean difference in interobserver agreement (Bland-Altman statistics) ranged from -0.25 mm² to 0.27 mm², with no significant difference between senior and junior readers. Multifocal lesion or foveal involvement in atrophy was not the cause of disagreement. CONCLUSION: Region Finder is a reliable tool for the identification and quantification of geographic atrophy in patients with age-related macular degeneration, in a clinical setting even when performed by junior reader.