Abstract
Abstract: :
Purpose: To analyse ICG pre–injection fluorescence and to differentiate pseudofluorescence and autofluorescence. Methods: Patients undergoing Indocyanine green (ICG) angiography for macular pathology were included. Imaging was performed with a confocal scanning laser ophthalmoscope (HRA I) and included red–free imaging, infrared (IR) imaging (820 nm), imaging in fluorescein angiograpy mode (488 nm excitation, barrier filter 505 nm, cut off 10–6) and ICG–angiography mode ( excitation 790 nm, barrier filter 805 nm, cut off 10–6) . In cases where infrared pre–injection fluorescence was observed, a series of 5 to 9 images were averaged and compared with native IR–images and related to fundus structures. Results: We included 270 consecutive patients. Pre–injection fluorescence was observed in 46 eyes (17.04 %. ) which was graded as weak in 22 eyes (8.15%) and as strong in 24 eyes (8,9%). Of those patients showing weak pre–injection fluorescence, 7 had dry AMD with pigment clumping, 1 had chloroquin maculopathy, 2 had nevi, 10 had exsudative AMD, 2 had small hemorrhages. In eyes with strong pre–injection fluorescence 14 had exudative AMD with choroidal neovascularisation surrounded with a pigmented ring, 8 had older greyish hemorrhages, 2 had a nevus. Exept for 1 case, comparison with IR–images revealed that retinal structures showing pre–injection fluorescence were highly reflective in IR–images, indicating that pre–injection fluorescence derives from leakage of the barrier filter. Haemorrhages which showed strong pseudofluorescence did in some cases show autofluorescence at 490 nm. Autofluorescence at 490 nm was increased in areas of pseudofluorescence in 21 of 31 AMD patients and in none of the nevi patients. Pseudofluorescence signals were not evident after ICG was injected. Conclusions: In our patient group we found a lower rate of pseudofluorescence compared to previous studies. The difference is most likely a result of technical differences in the imaging systems used. In our study eyes showing an IR pre–injection fluorescence had a strong infrared reflectancy at identical fundus location. Therefore, we could show that exept for 1 case at 790 nm not autofluorescence but pseudofluorescence can be observed. with AMD, particularly if the lesions were pigmented. The composition of pigmented lesions can be subtyped by their signal pattern seen at various wavelenghts.
Keywords: imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • retina • macula/fovea