"From the beginning, it was not totally clear if near-infrared fluorescence imaging is possible in a rat animal model, nor was it clear if retinal cross-sections could be evaluated ex vivo at high-resolution with the confocal scanning laser ophthalmoscope," explains Dr Steffen SchmitzValckenberg (a private lecturer at the Department of Ophthalmology, University of Bonn, Germany).

As a result, Dr Schmitz-Valckenberg and his team devised an investigation, featured in the British Journal of Ophthalmology, evaluating nearinfrared (NIR) autofluorescence (AF) in patients presenting with geographic atrophy (GA) secondary to agerelated macular degeneration (AMD). The paper also used in vivo and histological analysis to study the origin of the signal in rats and a human donor eye.

Time is of the essence, according to Dr SchmitzValckenberg. He stressed that cutting down observational time and reducing the number of patients is advantageous in the early identification of high-risk GA, "Currently, the identification of so-called 'rapid-progressors' is important for the design of interventional trials. While it is well known that the rate of progression shows extensive differences in between individuals, the exact reasons are not entirely understood. Depending on potential side-effects, it would make common sense only to treat highrisk patients, while a specific therapy might be too risky when only expecting a relatively low disease progression with preservation of vision for years."

Conducting the confocal laser study

"Several other groups have published finding of NIR AF in other retinal disease in the last years, including hereditary retinal disease and central serous retinopathy."

The experiment involved using confocal scanning laser ophthalmoscopy in vivo imaging. The excitation was set at 488 nm and emission was 500–700 nm and for NIR excitation was 790 nm and emission was >810 nm.

AF was conducted in 21 eyes of 18 GA patients. Dr Schmitz-Valckenberg claims, "The confocal scanning laser ophthalmoscope optimally addresses the challenges of reflectance and fluorescence imaging over large retinal areas — both in the human and rat animal model. In this study, the Heidelberg Retina Angiograph 2 was used. No other system was evaluated."

"It makes a lots of sense that the use of different excitation and emission wavelengths would lead to the detection of fluorescence from a different source. The current study adds data to the current understanding that blue AF is mainly originated from lipofuscin while NIR AF is derived from melanin."

Albino and pigmented rats underwent in vivo and post-mortem imaging with the same device. An additional customised magnification lens was used to image Cryostat-prepared retinal cross-sections. The last phase of the experiment involved recording a cross-section of a 49-year-old human donor eye.

Emerging research tool

"The investigation confirmed that the anticipated results matched the actual results. Previous studies indicated that the near-infrared fluorescence signal was derived from the choroid and retinal pigment epithelium. At the same time, it was assumed that the signal strength was relatively low compared to other retinal imaging modalities," said Dr SchmitzValckenberg.

The data suggests that selective imaging of melanin has a lot to offer when it comes to new applications in retinal diseases. Dr Schmitz-Valckenberg explained, "Alterations in metabolisms of melanin are assumed in many different retinal disorders. Currently, NIR AF imaging can be regarded as an emerging research tool."

This is an exciting revelation in the field of retinal imaging because the study has emphasized the claim that melanin is the main source of NIR AF in the healthy retina. Dr Schmitz-Valckenberg said, "Increased NIR AF intensities within the junctional zone in GA eyes could represent a significant build up of melanolipofuscin. This could certianly pinpoint and reflect disease acitivty."