Purchase this article with an account.
D. V. Palanker, C. Sramek, H. Nomoto, Y. M. Paulus, J. Brown, P. Huie; Computational Model of Retinal Photocoagulation. Invest. Ophthalmol. Vis. Sci. 2009;50(13):222.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Despite the extensive use of retinal photocoagulation in clinical practice, the dependence of the lesion size and retinal rupture on laser parameters (beam size and shape, pulse duration, and power) are poorly understood and characterized. We developed a computational model that can predict the width and depth of retinal lesions for many laser parameters, as well as the corresponding safe therapeutic range. This model can provide guidance for retinal therapy determining the extent of thermal damage in tissue, and the number of lesions required to coagulate certain fraction of the retinal surface area.
Absorption of 532 nm radiation was measured in the pigmented retinal layers in freshly enucleated rabbit eyes, including variability in RPE cell size and pigmentation, as well as beam shape and power transmission. Retinal coagulation and rupture (vaporization) thresholds were measured in-vivo and in-vitro with pulse durations varying from 1 to 200 ms. RPE damage thresholds were assessed using a fluorescent viability assay. This data was used to construct and validate an Arrhenius model of retinal photocoagulation using the finite-element computational package COMSOL 3.4.
The peak temperature at rupture threshold was in the range of 180 - 190 ºC for pulse durations from 1 to 200 ms. An Arrhenius damage integral with activation energy of 340 kJ/mol (3.5 eV) properly described the measured RPE damage thresholds. The predicted lesion width showed good quantitative agreement with histological data over a wide range of pulse parameters, corresponding to clinical grades of retinal lesions ranging from barely visible to intense. The use of a ring-shaped beam profile eliminates the central hot spot characteristic of top-hat and Gaussian beams, and is expected to double the safe therapeutic range for short (from 1 to 10 ms) pulse durations.
A computational model of retinal photocoagulation can be used to determine the required number of laser exposures needed to treat a given retinal area with arbitrary laser parameters. This allows for the proper comparison of the clinical efficacy of various laser treatments, while assuring that the same retinal area is coagulated at all settings. An optimized ring-shaped beam profile can allow for safe photocoagulation with pulse durations in the range of a few milliseconds, which can help to further reduce pain and improve spatial confinement of the thermal effects in the retina.
This PDF is available to Subscribers Only