Abstract
Purpose :
The purpose of this study is to compare the precision and agreement of macular and optic nerve head (ONH) parameters from two models of Cirrus OCT in healthy population. The newer model Cirrus (6000) achieves higher acquisition speed using a camera capture rate of 100 kHz compared to the 27 kHz of the older model Cirrus (5000).
Methods :
In this prospective study, Macular 512x128 and Optic Disc 200x200 scans were obtained from eligible participants with healthy eyes. Three OCTs each from CIRRUS® OCT 5000 and 6000 (ZEISS, Dublin, CA) models were used. Three scans/operator/device were performed for each subject. The eye/operator/device order were randomly assigned, and analyses were performed using 11.5.2 software. The following were assessed: 1) macular scans: ganglion cell thickness (GCT) and macular thickness (MT) in nine regions from Early Treatment of Diabetic Retinopathy Study (ETDRS); 2) optic disc scans: ONH parameters, retinal nerve fiber layer (RNFL) thickness. All scans were monitored for quality in real-time as well as during the data qualification process by an expert grader. Analysis of variance (ANOVA) was performed to evaluate precision and Bland-Altman and Deming regression were performed to assess agreement. Variance components were reported as standard deviations (SDs).
Results :
A total of 25 eyes from 25 healthy subjects were included in the study. 225 macular scans and 214 optic disc scans were determined to be valid from both Cirrus models. Repeatability and reproducibility CV%s for all the parameters are low (<10%) and clinically acceptable. For MT, GCT, ONH and RNFL, repeatability and reproducibility SD ratios (Cirrus 6000/Cirrus 5000) were in the range 0.64-1.17, 0.79-1.23, 1.30-1.71 and 0.80-1.30 (Table 1). Slightly higher repeatability and reproducibility SD ratios in ONH parameters are due to small SD values exaggerating any differences. Table 2 shows the results of Bland-Altman analyses and Deming regression that shows good agreement between both Cirrus models.
Conclusions :
Repeatability and reproducibility results shows high precision in both Cirrus models. Study results from healthy population show good agreement between two Cirrus models. This suggests Cirrus 6000 model can be used in clinic in a similar way as the Cirrus 5000.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.