One year progression of geographic atrophy assessed with polarization-sensitive optical coherence tomography and autofluorescence imaging

Abstract

 

Purpose

 

To quantify the growth of geographic atrophy (GA) using polarization-sensitive optical coherence tomography (PS-OCT) during a follow-up time of 12 months in comparison to autofluorescence (AF)-imaging.

 

Methods

 

Twenty eyes of 13 patients (8 female, 5 male; mean age ± SD = 76,4 ± 7,4 years) diagnosed with unilateral or bilateral GA were included in the study and imaged with PS-OCT and AF every three months in order to evaluate the progression of the GA-area. PS-OCT segments the RPE based on its depolarizing tissue-specific properties. The device used for this study is capable of recording 3D datasets of the human retina at a scanning speed of 70.000 A scans/s with scanning angles of up to 40x40°. Using proprietary software, GA area was segmented in a semiautomatic fashion and GA area sizes were calculated in square-pixels. AF-images were acquired with Heidelberg Spectralis HRA + OCT® using image averaging. The GA areas were quantified (in square-mm) using the region finder software, an algorithm embedded in Spectralis® that semiautomatically segments GA areas in AF images. Mean progression of GA areas as compared to the baseline size was calculated at 3, 6, 9 and 12 months using both imaging modalities. Unpaired t-tests were performed to test for a difference between region growth measured with PS-OCT and AF, respectively.

 

Results

 

A total of 86 PS-OCT volume scans with a scanning angle of 30x30° (each consisting of 1024 × 250 A-scans) and 86 averaged AF-images (30x30°) were evaluated for GA size and compared. There was no statistically significant difference in GA-progression between AF and PS-OCT at any time during follow-up (p= 0.719 at 3 months, p= 0.957 at 6 months, p= 0.763 at 9 months, p= 0.803 at 12 months). Mean GA-progression was 1.21 ± 0.41 and 1.16± 0.39 at 3 months, 1.33± 0.62 and 1.32± 0.86 at 6 months, 1.67 ± 1.57 and 1.51 ± 1.32 at 9 months and 1.72 ± 1.63 and 1.59 ± 1.66 at 12 months in PS-OCT and AF, respectively.

 

Conclusions

 

PS-OCT has been shown to have comparable performance in segmentation and quantification of GA areas as AF, the current gold-standard in imaging and quantification of GA. Additionally, PS-OCT allows for a detailed analysis of all layers of the retina (e.g. microstructural changes of the neuroretina at the junctional zone of the GA), thus providing superior information to two-dimensional imaging modalities like AF.

 

Keywords: 412 age-related macular degeneration • 550 imaging/image analysis: clinical • 701 retinal pigment epithelium