Repeatability and reproducibility of optical coherence tomography (OCT) measurements on CIRRUS 6000 compared to CIRRUS 5000 in retinal diseased eyes

Abstract

Purpose : The objective of this study was to determine the repeatability and reproducibility of CIRRUS HD-OCT macular measurements on ZEISS CIRRUS^® OCT 6000 (ZEISS, Dublin, CA) as compared to ZEISS CIRRUS^® OCT 5000 (ZEISS, Dublin, CA) on eyes with retinal disease.

Methods : Subjects with a variety of retinal disease such as ERM, AMD, and mild to severe PDR were scanned with 3 CIRRUS 5000 (C5000) and 3 CIRRUS 6000 (C6000) devices with the Macular Cube 512x128 scan. For each subject, 3 scans/operator/device were performed. Each operator was assigned a specific C5000 and C6000 unit. The Macular Thickness (MT) in the 9 ETDRS regions and 8 Ganglion Cell Thickness (GCT) parameters were assessed. GCT is the sum of ganglion cell + inner plexiform layer thicknesses. Variance components (including device/operator configurations, DevOp), coefficient of variation (%CV), and R&R limits estimated from Analysis of Variance (ANOVA) were produced for all measurements. Variances were reported as Standard Deviations (SDs). Agreement was evaluated by Bland-Altman plots and Deming regression.

Results : The study included 29 eligible subjects with retinal diseases. MT displayed excellent %CV with all values below 2%. For GCT, all %CV were ≤ 8.60%. The summary results are presented in Table 1. Bland-Altman plots presented generally good limits of agreement for all parameters and Deming regression results show good agreement of the Macular Cube 512x128 values between the two devices (Figure 1).

Conclusions : The study analyses demonstrate repeatability and reproducibility limits of GCT and MT across both devices are clinically acceptable. GCT analysis performed on retinal pathology may cause an increased variability in these measurements. As retinal lesions are more susceptible to GC segmentation errors, clinicians should be aware of macular pathology affecting this analysis. Results present overall good agreement across all Macular Cube 512x128 parameters between the C6000 and C5000.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.