First Author: M.Oishi JAPAN
Co Author(s): A. Oishi M. Lindner M. Pfau S. Schmitz-Valckenberg F. Holz M. Fleckenstein
Back to previous
Increased fundus autofluorescence (FAF) signals in the border zone of geographic atrophy (GA) secondary to age-related macular degeneration (AMD) and specific abnormal FAF patterns have been shown to impact GA progression over time. However, the underlying structural correlates as well as the cellular and molecular mechanisms responsible for localized increased FAF signal intensity are unclear. Herein, we investigate structural correlates of lesions with increased FAF on spectral domain-optical coherence tomography (SD-OCT) in the perilesional area of GA secondary to AMD.
Prospective longitudinal study
Serial FAF images of 181 eyes with GA of 134 patients participating in the Directional Spread in Geographic Atrophy (DSGA) study (NCT02051998) were screened for increased FAF spots which had developed during the review period. In the current study, an increased FAF spot was defined as a well-demarcated hyperautofluorescent area 100 -500 µm in diameter. Included were only spots for which serial OCT scans of exactly the area were available. The mean thickness and the mean reflectivity of the retinal pigment epithelium (RPE)-basal lamina complex in corresponding OCT scans were measured in the width of 100 µm using the ImageJ software and were compared between the time-points before and after the appearance of increased FAF. As for reflectivity, the mean reflectivity of the entire retina was measured as internal reference, and the ratio of the mean reflectivity of the RPE-basal lamina complex to that of the entire retina was used as a ‘reflectivity ratio’. In addition, the integrity of the external limiting membrane (ELM) and the ellipsoid zone (EZ) in corresponding OCT scans were categorized in ‘continuous’ or ‘disrupted’, respectively, and were compared before and after the appearance of increased FAF.
A total of 36 spots in 15 eyes which developed de novo during the review period were included. Mean age of the participants was 78.9 ± 5.6 years, and mean interval between the time-points before and after the occurrence of increased FAF spots was 17.4 ± 8.0 months (range 6 - 36 months). Analysis of the corresponding OCT images revealed an increase in RPE-basal lamina complex thickness (31.8 ± 7.3 µm to 42.1 ± 11.9 [P < 0.001]) and reflectivity (reflectivity ratio: 1.42 ± 0.11 to 1.54 ± 0.27 [P = 0.009]) corresponding to an increased FAF signal. There was no significant association between the two parameters (R2 = 0.011). Disruptions of the ELM and EZ were observed more frequently when increased FAF was observed (ELM disruptions: 0/36 vs. 9/36, P = 0.03; EZ disruptions: 6/36 vs. 33/36, P < 0.001).
The findings of this study indicate that de novo occurrence of spots with increased FAF in eyes with GA due to AMD correlates with both increase in reflectivity and thickness of the RPE-basal lamina complex. In addition, disintegration of outer retinal structures on OCT images correlate with newly occurring spots of increased FAF. Thus, structural changes underlie focal enhancements of FAF. Co-registration of both imaging modalities, i.e. FAF and SD-OCT-imaging with exact topographic alignment add to the understanding of the phenotypic variability in FAF signals in the context of dry AMD. Additional longitudinal studies are warranted to further elucidate the natural history of the observed FAF changes.