Abstract
Purpose :
To evaluate the accuracy of multicolour imaging compared to standard colour fundus photography (CFP) in differentiating AMD from normal eyes, and in detecting features of PCV.
Methods :
In a prospective study of 50 consecutive patients presenting with PCV or AMD, standardized multimodal imaging (CFP, multicolour imaging, fluorescein and indoyanine green angiography) was performed. PCV was diagnosed using specific diagnostic criteria, with ICGA as the gold standard. CFP and multicolor images were graded using standardized grading protocols to determine sensitivity, specificity, positive and negative predictive values (PPV and NPV) in differentiating AMD from normal eyes, and in detecting features of PCV.
Results :
The mean age of the patients was 70.0 years. Of 100 eyes, 44 had PCV, 11 had neovascular AMD, 21 non-neovascular AMD and 23 were normal. Of 44 eyes, polyps appeared as dark green oval lesions in 39 (88.6%) using the multicolour channel, while the branching vascular network (BVN) appeared as mottled grey lesions in 16 (36.4%) using the infrared channel. Multicolor had superior specificity (73.9% vs. 52.2%) and NPV (94% vs. 85.7%) compared to CFP for detecting all types of AMD. The sensitivity was similar for both multicolor and CFP (97.7% vs. 97.4%). For the detection of PCV, multicolour had higher sensitivity (86.4% vs. 59.1%) and NPV (89.3% vs. 74.3%) compared to CFP. In contrast, the specificity (89.3% vs. 92.9%) and PPV (86.4% vs. 86.7%) were similar between the two. PCV lesions were best visualized on the infrared multicolour images. Using BVN as a parameter, infrared imaging had very high specificity (96.6%) and PPV (88.9%) for detecting PCV.
Conclusions :
Multicolour imaging is superior to CFP in differentiating AMD from normal eyes and detecting features of PCV. The presence of BVN on infrared imaging and dark green oval lesions should alert ophthalmologists to the presence of PCV.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.