Purchase this article with an account.
C.N. Keilhauer, J. Fischer, J. Mlynski, A. Stangl, F.C. Delori; Confocal Infrared Autofluorescence Imaging in Normal Subjects and AMD Patients . Invest. Ophthalmol. Vis. Sci. 2005;46(13):1394.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: The ocular fundus exhibits a distinct, albeit weak, infra–red autofluorescence (Picollino, 1996). We used confocal scanning laser ophthalmoscopy to visualize this autofluorescence. Our goal was to identify which pigments are responsible for the fluorescence and to explore the clinical merits of this image modality. Methods: Infra–red autofluorescence (IR–AF) images were recorded in 34 normal subjects (ages: 10–80 years) and in 68 patients with AMD and other retinal diseases (ages: 21–87 years). The IR–AF images were obtained with a standard HRA (Heidelberg Retinal Angiograph) using the excitation power and wavelength (787 nm) and the detection filter (>800 nm) normally used for ICG angiography. Images were the average of 18 frames. For all subjects we also recorded 488–nm AF (BL–AF), confocal IR (820 nm), and color images. Results: The AF efficiency was estimated for 2 subjects (40–46 years old) to be 220 times lower for IR–AF than for BL–AF. Given the higher sensitivity and higher retinal exposure in the IR, the recorded signal is about 10 times smaller for IR–AF than for BL–AF. The most striking feature of the IR–AF images was an elliptical area of high fluorescence centered on the fovea. Its mean horizontal diameter was 8.0°±2.5° (2 observers), not significantly different than the dark area seen on BL–AF images (area of higher RPE melanin). The ratio of grey–levels at the fovea to that at 2–3° from the temporal disk was 1.47±0.14. This ratio showed a tendency to decrease with age (p=0.05). Analysis of IR–AF contributions from the stratified fundus layers in normal and pathological retinas (e.g. choroidal crescent) suggested that the RPE is in large part responsible for the IR–AF signal, but that the choroidal layers and the sclera also contribute to the overall signal. Analysis of various fundus pathologies in AMD patients revealed high IR–AF contrast at the sites of hyperpigmentation, some (but not all) lipofuscin accumulations, old hemorrhages and certain neovascular membranes. The distribution of IR–AF did not always correspond to that of the BL–AF, but was often similar to the distribution of high reflecting features of the IR–images. Conclusions: Melanin, and perhaps a component of lipofuscin, contribute substantially to the infra–red autofluorescence. Other fluorophores (porphyrin, collagen) can not be excluded at this stage. Confocal IR–AF with the HRA may provide a new modality to visualize pathological features of the RPE and the choroid in AMD.
This PDF is available to Subscribers Only