Objective: To quantify the spatial extent of pigment epithelial detachment (PED) associated with age-related macular degeneration (AMD) using a rapid, noninvasive method.
Design: Prospective, cross-sectional study.
Participants: The authors tested 32 eyes of 21 patients (13 women and 8 men) with AMD 56 to 91 years of age (mean = 72.8 years). The authors retested seven eyes of six patients.
Intervention: Three-dimensional imaging and quantification of PED were performed in each subject using confocal infrared imaging (790 nm) with a Topographic Scanning System (TopSS). The data consisted of a series of 32 images within a 3-mm-depth range requiring 0.9 second. Three-dimensional calculations were made from the series.
Main outcome measure: Height, diameter, area, volume, and slope of each PED were obtained with two calculation methods. The Ellipse Method used a graphics tool to draw elliptical borders circumscribing the region of interest in the image. The software automatically calculated the values for all parameters for the region inside the ellipse, with the retinal reference plane adjusted to match the height of the surrounding retina. The User-Defined Region Method differed in that the region of interest was drawn manually.
Results: PEDs were easily detected in all patients using the TopSS. Maximum height of all PEDs above the reference plane ranged from 0.204 to 1.818 mm (mean = 0.57 mm). The diameter was 0.501 to 5.151 mm (mean = 2.711 mm), area was 0.179 to 20.402 mm2 (mean = 6.585 mm2), and volume was 0.012 to 13.981 mm3 (mean = 2.173 mm3). Intraobserver variability was low, with correlations between first and second measurements for the first visit ranging from r = 0.906 to 0.997 for slope and diameter, respectively.
Conclusions: Three-dimensional analysis with the TopSS provides objective outcome measures not obtainable with typical clinical methods such as fundus photography and angiography. Height and volume are crucial in determining whether neovascularization is worsening or persistent after photocoagulation. More longitudinal data are needed to determine whether tomographic data reduce the need for angiography. Unique to scanning laser tomography, exudative features were imaged at different depths.