Non–Invasive Optical Monitoring of Choroidal Reaction to Retinal Laser Therapy

Abstract

Abstract: : Purpose: Hyperthermic therapies such as Transpupillary Thermal Therapy (TTT) can be efficient only in a very narrow therapeutic window: above the threshold of expression of the heat shock proteins (43^oC) but below the threshold of irreversible cellular damage (48–50^oC). Large variability in the retinal pigment epithelium (RPE) pigmentation and choroidal perfusion results in significant variation of retinal temperature, making TTT unpredictable and inefficient. Direct monitoring of the laser effects on tissue during the procedure is required. Methods: TTT was performed on rabbits (790nm laser wavelength, 60 seconds, 0.86mm retinal spot) at power levels of 10–100 mW. Simultaneously the fundus was illuminated with linearly polarized broadband white light (7mW, 2.2 mm spot) and the spectrum of light (400–750nm) back–scattered from the central part of the laser spot (0.5mm diameter) was measured in parallel and cross polarizations. In addition, visibility of the laser–induced lesion was evaluated during the procedure. The effect of laser heating on blood vessels was observed on the medullary ray of the rabbit retina. Results: Strong vasoconstriction of retinal blood vessels was observed during the laser exposure. Increased light scattering at wavelengths shorter than 600nm was observed at laser irradiance 70% below the visible retinal damage threshold. Spectral changes were maximal at the two absorption peaks of oxyhemoglobin: 540nm and 570nm. Scattering light intensity at these peaks increase up to 20% in the invisible lesions and up to 40% in the minimally–visible lesions. The characteristic time constant of this spectral response is about 20 seconds. Sensitivity of the cross–polarized detection is twice that of the parallel polarization. Spectral changes did not completely disappear after the laser exposure even below the levels of visible retinal damage. These laser–induced spectral changes, as well as vasoconstriction of the retinal vessels were not observed in a euthanized rabbit. Conclusions: The effects observed are detectable at exposure levels corresponding to TTT and PDT. Since rabbit retina is avascular the relative increase in scattering at the wavelengths corresponding to oxygenated blood is most probably due to choroidal vasoconstriction. This effect could be used for non–invasive monitoring of the tissue reaction to heating during TTT treatment.

Keywords: choroid • laser • blood supply