Do subretinal drusenoid deposits have a spectral fingerprint?

Abstract

Purpose : Retinal imaging using near-infrared (NIR) reflectance at wavelength ~820 nm provides high sensitivity for detecting subretinal drusenoid deposits (SDD), also called pseudodrusen. We imaged well-characterized SDD with lights of different wavelengths in the NIR region, using high-resolution multi-spectral adaptive optics scanning laser ophthalmoscopy (AOSLO).

Methods : The AOSLO was equipped with a supercontinuum laser (NKT Photonics A/S, Denmark) that provides imaging lights over a spectral range of 650 – 1200 nm. An acousto-optical tunable filter was used to tune a single wavelength for spectral imaging. AOSLO was used to image the macula of a patient with age-related macular degeneration at 8 wavelengths (680 nm, 700 nm, 720 nm, 750 nm, 780 nm, 800 nm, 820 nm, and 840 nm). The progression stage of each individual lesion has been characterized yearly over a 5-year follow-up by multimodal imaging, using a 3-stage grading system for spectral domain coherence tomography. At the last visit, individual SDD lesions were examined by AOSLO to evaluate variations in reflectivity with different spectral light. A normalized spectral reflectivity of individual lesions was defined by the ratio of reflectivity in the lesion core to reflectivity in nearby photoreceptor mosaic that was undisturbed by SDD. Twenty-two stage 3 SDD were measured.

Results : High resolution AOSLO images acquired at different wavelengths allowed SDD spectral reflectivity assessed at individual lesion level. Data from the subject indicate SDD reflectivity in the NIR (680 - 840 nm) region may vary with wavelength, with greater reflectivity at 720 nm and 820 nm (Figure 1). However, due to small sample size, variability is large.

Conclusions : We demonstrate that AOSLO multi-spectral imaging can be used to characterize the spectral reflectivity of SDD, with the goal of generating a spectral fingerprint of SDD in the NIR region. Ongoing research will investigate more lesions in more subjects. This approach, if extended to longer wavelengths and combined with histology and mass spectrometry of donor eye samples, may enable the molecular bases of SDD to be studied in vivo.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.

 

Figure 1 SDD spectral reflectivity assessed by multi-spectral AOSLO. The error bars indicate ±1 standard deviation.