In recent years, powerful imaging and diagnostic technology has revolutionized and redefined clinical retina practice. Fundus autofluorescence (FAF) is becoming an increasingly important diagnostic tool in a spectrum of retinal diseases, particularly age-related macular degeneration (AMD). This is a fast-emerging, noninvasive technique that permits the topographic mapping of lipofuscin distribution in the retinal pigment epithelial (RPE) cell monolayer and provides information additional to conventional mapping techniques. Excessive accumulation of lipofuscin commonly represents a negative pathogenetic pathway in AMD, as well as other hereditary retinal diseases.1

FAF imaging has been shown to be useful with regard to understanding of pathophysiologic mechanisms, diagnostics, phenotype-genotype correlation, identification of predictive markers for disease progression, and monitoring of novel therapies. FAF imaging gives information above and beyond that obtained by conventional imaging methods, such as fundus photography, fluorescein angiography, and optical coherence tomography. Its clinical value, coupled with its simple, efficient, and noninvasive nature, is increasingly appreciated.2

Clinically, the Spectralis (Heidelberg Engineering, Vista, CA) models are combined confocal scanning laser ophthalmolscopy (cSLO) fundus imaging and spectral-domain (OSD) OCT BluePeak imaging, four of which are enabled with blue laser autofluorescence capabilities.

APPLICATIONS OF FUNDUS
AUTOFLUORESCENCE
FAF has a variety of useful applications in the clinic. First, it is helpful with making a differential diagnosis of macular degeneration. There are a number of “mimicker” diseases, such as adult-onset foveal macular dystrophy or pseudovitelliform degeneration, in which optical coherence tomography (OCT) images and fluorescein angiography (FA) can appear similar to wet AMD. Because the material that has been deposited under the retina is very hyperfluorescent, however, using FAF on these eyes shows a tremendous amount of autofluorescence, which is not present with standard AMD. Anything that eliminates misdiagnosis has, without question, utility.

A second application for FAF in the clinic is in following patients with geographic atrophy (GA). The literature supports the notion that increased rim autofluorescence is associated with an increased rate of expansion of GA. Brar et al3 correlated high-resolution OCT images to FAF imaging to study the changes in appearance of the margins of GA and found that spectral-domain SD-OCT provided in vivo insight into the pathogenesis of GA and its progression (Figure 1). Visualization of reactive changes in the retinal pigment epithelial cells at the junctional zone os correlated with increased FAF secondary to increased lipofuscin; together, these methods may serve as determinants of progression of geographic atrophy.

Additional studies have corroborated this. Although there are no effective treatments for GA, the information does help in counseling patients. In addition, clinical studies targeting GA, of which there are now several, almost universally require FAF as one of the imaging tools in the trials. This requirement results because FAF helps identify fast progressers and because there is some thinking that the area of GA can be better quantified with FAF than with fundus photography.

A third utility for FAF is to increase our understanding of choroidal neovascularization (CNV) in patients with AMD (Figure 2). A study by Dandeker et al4 showed that preserved autofluorescence in subjects with recent-onset CNV indicates viable retinal pigment epithelium initially, which has implications for visual prognosis. Decreased autofluorescence in subjects 1 to 6 months after diagnosis or with late-stage CNV indicates loss of RPE and photoreceptors.

Although these data are more speculative, identifying patterns of autoflorescence within the fovea of patients receiving anti-vascular endothelial growth factor (VEGF) therapy may be helpful in assigning a treatment regimen. If there is complete lack of autofluorescence, suggesting that the RPE is extremely sick or absent, there is the suggestion that the likelihood of vision improvement in those patients is diminished. If a normal autoflorescence pattern is maintained, there are some data to suggest that that patient has the capacity to improve vision. Again, this is speculative, but because we do not know whose vision improves and whose does not, and how to titrate our anti-VEGF therapies, this might prove to be a useful modality for following patients undergoing Anti- VEGF therapy for CNV.

The most cutting-edge investigations of FAF involve defining recurring patterns of autofluorescence in early AMD patients. An important advancement in studying the particular patterns of autofluorescence within AMD patients is possible now that an automated eye-tracker allows repeated OCT scans to be made of the exact same spot on the eye. The Spectralis HRA+OCT (Heidelberg Engineering, Vista, CA) contains an eye-tracking system that compensates for eye movement during the scan and results in multiple benefits. First, it allows multiple scans of the same area, producing a mean image that has much greater detail than a single image. Second, it allows precise correlation in changes of lipofuscin patterns on the retina. Whereas previously we studied greater or lesser quantities of autofluorescence, now specific names are being assigned to autofluorescence patterns— speckled, deflect, and butterfly.5 The particular pattern of autofluorescence seen in a patient with early AMD might predict if they end up with GA or CNV. Although at this point, these types of studies require much more research, FAF could be a powerful adjunct to the way that patients with high-risk dry AMD are managed.

SUMMARY
Data from the trials currently under way should add to the existing data correlating GA and CNV with quantities of autofluorescence, supporting FAF as an indispensable diagnostic tool in the clinic. This will make experience with FAF images essential. The variability in generating FAF images is operatordependent, and there is a learning curve. An experienced operator will be more successful in generating good images, analyzing the images, and following patients. Having the ability to create FAF images and the experience to analyze them will allow the practitioner to determine which patients should be followed and which patients should be treated.

Karl G. Csaky, MD, PhD, is the Director, Sybil B. Harrington Molecular Ophthalmology Laboratory, at the Retina Foundation of the Southwest, Dallas. Dr. Csaky states that he is a consultant for Heidelberg Engineering. He may be reached at +1 214-363-3911 ext. 137; or via e-mail at kcsaky@retinafoundation.org.

  1. Schmitz-Valckenberg S, Fleckenstein M, Scholl HP, Holz FG. Fundus autofluorescence and progression of age-related macular degeneration. Surv Ophthalmol. 2009;54(1):96–117.
  2. Schmitz-Valckenbery S, Holz FG, Bird AC, Spaide RF. Fundus autofluorescence imaging: review and perspectives. Retina. 2008;28(3):385–409.
  3. Brar M, Kozak I, Cheng L, t al. Correlation between spectral-domain optical coherence tomography and fundus autofluorescence at the margins of geographic atrophy. Am J Ophthalmol. 2009;148(3):439-44. Epub 2009 Jul 9.
  4. Dandeker SS, Jenkins SA, Peto T, et al. Autofluorescence imaging of choroidal neovascularization due to age-related macular degeneration. Arch Ophthalmol. 2005;123(11):1507–1513.
  5. Bindewald A, Bird AC, Dandekar SS, et al. Classification of fundus autofluorescence patterns in early age-related macular disease. Invest Ophthal Vis Sci. 2005;(9)46:3309.