Optical Coherence Tomography grading of AMD lesions

Abstract

 

Purpose

 

Optical coherence tomography (OCT) has been widely used in clinical settings. In epidemiological ophthalmological research the OCT has not been implicated on a large scale. With this study we aim to investigate the merit of OCT grading in addition to fundus photograph grading for age-related macular degeneration (AMD).

 

Methods

 

Participants (55+ yrs) of the population-based Rotterdam Study underwent multiple extensive ophthalmic examinations including fundus photography and OCT. Fundus photographs were taken after pharmacological mydriasis with a digital 35° fundus camera (Topcon TRC 50EX with a Sony DXC-950P digital camera;0.44 megapixel) and were graded according to the International Grading System . 3D-horizontal macular OCT scans (6x6mm) were made using the Topcon OCT-2000 (SD-OCT, Topcon Corp, Japan). Scans were graded with a recently developed OCT-grading system by international experts. Inter-observer reliability was calculated using Cohen’s kappa, and differences in graded lesions were analyzed by McNemar test.

 

Results

 

We included N=154 eyes of 154 participants with signs of AMD for comparison of both grading systems. Drusen were graded significantly more often on fundus photographs than on OCT ( 89.0% vs 81.8%, P=0.013). In particular small drusen (< 125 µm) were not visible on OCT. Pigmentary changes were difficult to grade on OCT (63.0% vs 16.2%, P<0.0001). Frequency of geographical atrophy was comparable between OCT and fundus photographs grading (5.8% vs 5.8%, P=1.00). Neovascular AMD and fibrous scars, however, were graded more often on OCT (5.2% vs 7.1%, P=0.38 and 2.6% vs 3.9%, P=0.69 respectively). Intra-observer κ value for the OCT-grading ranged from 0.60 for drusen to 1.00 for geographic atrophy and neovascular AMD.

 

Conclusions

 

Early AMD signs were more visible and therefore better gradable on fundus photographs. Late AMD subtypes, especially neovascular AMD and fibrous scars, were visible in more detail on OCT, and therefore more often diagnosed as such. OCT is complementary to fundus photographs grading and an improvement for AMD grading in epidemiological studies.

 

Keywords: 412 age-related macular degeneration • 551 imaging/image analysis: non-clinical • 465 clinical (human) or epidemiologic studies: systems/equipment/techniques