Subfoveal Choroidal Thickness

Enhanced depth imaging OCT was conducted based on previously described methods. By using the Heidelberg Eye Explorer software version 1.5.12.0, the choroidal thickness was measured manually as previously described. Choroidal thickness was defined as the vertical distance from the hyperreflective line of the Bruch membrane to the innermost hyperreflective line of the chorioscleral interface ( Figure 1 ). Because we experienced a discrepancy in the choroidal thickness measurements between measurements based on 1:1-pixel image and those based on 1:1-μm image in certain cases, all the measurements were performed based on the latter. The subfoveal choroidal thickness was compared between the eyes with RAP and typical exudative AMD. Intragroup comparison of differences in subfoveal choroidal thickness between the first and the last visits was conducted for both the RAP group and the exudative AMD group. The subfoveal choroidal thickness of fellow eyes without exudative change also was compared between the 2 disease groups.

Representative optical coherence tomography images (1:1-μm setting) from (Top) an eye with typical exudative age-related macular degeneration and (Bottom) an eye with retinal angiomatous proliferation. Choroidal thickness (black arrow) is defined as the vertical distance from the hyperreflective line of the Bruch membrane to the outermost hyperreflective line.

Extent and Density of Drusen

Fundus photographs of the posterior retina (30 degrees in diameter centered at the foveal center) were obtained. The extent of distribution of intermediate drusen (63 to 124 μm in diameter) and soft drusen (≥125 μm in diameter) was estimated based on fundus photographs. Studied eyes were classified into 3 groups according to their extent of distribution of drusen in the fundus photographs: small extent group, less than one third of the extent of the fundus photograph; intermediate extent group, one third to two thirds of extent of the fundus photograph; and large extent group, more than two thirds of extent of the fundus photograph ( Figure 2 ). If no drusen were observed, the eye was classified as being in the small extent group. The extent of distribution of drusen was compared between the RAP group and the typical exudative AMD group.

Fundus photograph showing the extent of the area of drusen (white, dotted line) in eye with retinal angiomatous proliferation. This photograph is representative of the intermediate extent group.

The density of drusen in the fellow eye was determined by the horizontal and vertical crosshair enhanced depth imaging OCT scans through the fovea. Because the density of drusen could not be determined accurately in eyes with exudative change, only images of fellow eyes without exudative change were assessed. All images on OCT were converted to gray scale for better visualization and then were imported to Image J (freeware; National Institute of Health, Bethesda, Maryland, USA) for analysis. The entire length of the Bruch membrane within 3000 μm in diameter centered at the foveal center was measured using a segmented line tool. The total length of the Bruch membrane beneath the drusen, located within 3000 μm in diameter centered at the foveal center, also was measured using a segmented line tool ( Figure 3 ). In our pilot study, it was difficult to measure accurately the diameters of drusen that were smaller than 70 μm. Thus, we defined diameters of 70 μm or more as a measurable size, and the measurements were obtained only for measurable sizes of drusen. The density of drusen was calculated as the total length of the Bruch membrane beneath the drusen divided by entire length of the Bruch membrane, and the mean density of drusen obtained from both horizontal and vertical scans was used for analysis. The density of drusen was compared between the RAP group and the typical exudative AMD group. Correlation between the density of drusen and subfoveal choroidal thickness was estimated in the fellow eyes of patients with both RAP and typical exudative AMD. In addition, the diagnostic value of the extent of drusen and choroidal thickness for diagnosing RAP was analyzed.

Optical coherence tomography image in the fellow eye of an eye with retinal angiomatous proliferation showing method of estimating density of drusen. A white line indicated by a white arrowhead represents the total length of the Bruch membrane. Black segmented lines indicated by black arrows represent the length of the Bruch membrane beneath more than 70 μm in diameter of drusen. In this eye, the density of drusen was 0.236.

All the measurements were conducted by a single trained examiner (J.H.K.). Statistical analyses were performed with a commercially available software package (SPSS version 18.0 for Windows; SPSS Inc, Chicago, Illinois, USA). A test of normality was performed using the Shapiro-Wilk test. Values of baseline characteristics were compared using the Mann–Whitney U test and chi-square test. Comparison of subfoveal choroidal thickness between the 2 disease groups was performed using the Mann–Whitney U test. The subfoveal choroidal thickness and density of drusen of the fellow eyes were compared using the Mann–Whitney U test. The correlation between the density of drusen and subfoveal choroidal thickness in fellow eyes was estimated using the Spearman correlation analysis. Comparison of the subfoveal choroidal thickness between 2 time points was performed using the Wilcoxon signed-rank test. A comparison of the distribution of eyes among groups for the extent of drusen was performed using the Cochran-Armitage test. In addition, the diagnostic value and cutoff value of the extent of drusen and choroidal thickness were estimated using univariate logistic regression with receiver operating characteristic curve analysis. A P value less than .05 was considered significant.