We read with great interest the article describing treatment response of pigment epithelial detachment (PED) to anti–vascular endothelial growth factor (anti-VEGF) treatment in neovascular age-related macular degeneration (nAMD) by Cho and associates. Serous and fibrovascular PED are thought to be responsible for significant visual distortion, and characterization of optical coherence tomography (OCT)-derived PED height and morphology is critical for the development of possible treatment regimens. Development of OCT technology has improved the resolution of macular imaging; however, an understanding of repeatability of manual PED measurements is essential to distinguish true clinical change from measurement variability alone.

We recently conducted a prospective cross-sectional study to determine the intrasession repeatability of PED measurements derived from both Spectralis spectral-domain OCT (SDOCT) (Heidelberg Engineering) and DRI-OCT-1 swept-source OCT (SSOCT) (Topcon) in patients with subfoveal nAMD. This analysis used data from the DOCS (Distance of Choroid Study) at Moorfields Eye Hospital, London in 2014. The study was approved by the local ethics committee (NRES Committee South East Coast – Surrey). All patients underwent 3 Spectralis (macular volume 25 horizontal line scan, 512 A-scans per B-scan in enhanced depth imaging mode) and 3 DRI-OCT-1 (12 × 9-mm raster scan centered on the fovea) scans in the same imaging session. Individual line scans were assessed for the presence of PEDs and the highest PED of each OCT was determined. Manual measurements of this PED height were made using the calipers on the proprietary software measuring from the RPE to the Bruch membrane. Repeatability was calculated using methods described by Bland and Altman.

Thirty-six eyes of 36 patients with nAMD were included in the study. The mean (± standard deviation) age was 73.9 (± 7.2) years and mean lesion size was 1.15 (± 0.8) mm ² . There were 5 serous (13.9%), 25 fibrovascular (69.4%), and 4 drusenoid (11.1%) PEDs. SDOCT- and SSOCT-derived mean PED height was 223.7 (± 5.7) μm and 223.2 (± 13.5) μm, respectively. OCT-derived mean PED height taken with manual measurements from the 2 platforms correlated well ( P = .935, paired t test, 95% limits of agreement −13.41 to 13.89). The intrasession coefficient of repeatability (CR) of SDOCT PED height was 34.6 μm (95% CI 33.6–35.6) and SSOCT PED height was 46.9 μm (95% CI 45.5–49.7).

This study confirms that similar PED height measurements can be derived from both SDOCT and SSOCT in patients with nAMD. However, it suggests that SDOCT-derived mean PED height measurements are more repeatable than SSOCT. Furthermore, we report that a change of more than 34.6 μm and 46.9 μm in PED height when measured with Spectralis SDOCT and DRI-OCT-1 SSOCT, respectively, is required to suggest clinical change due to alteration in disease activity rather than measurement variability alone. Clinicians should be aware of the repeatability of OCT platforms when assessing patients with nAMD, and these repeatability estimates can help interpret studies of treatment of PED in nAMD and inform treatment decisions with anti-VEGF agents.