Intrasession, intersession, and interexaminer variabilities of retinal nerve fiber layer measurements with spectral-domain OCT

Abstract

Purpose: To evaluate the intrasession, intersession, and interexaminer variabilities of retinal nerve fiber layer measurements (RNFL) with spectral-domain optical coherence tomography (OCT).

Methods: A total of 32 healthy individuals and 34 patients with chronic glaucoma underwent RNFL measurements with the Cirrus HD-OCT Model 4000 (Carl Zeiss Meditec, Dublin, CA, USA) 5 times during the same sitting by one examiner to assess intrasession variability. The same examiner performed RNFL measurements in the same patients on 5 different days to assess intersession variability. A second examiner performed RNFL measurements in the same patients to assess interexaminer variability. The coefficients of variation and intraclass correlation coefficients were obtained for the following parameters: average thickness, quadrant thickness, and Clock hour thickness measurements.

Results: Intrasession variability: In patients with glaucoma, coefficients of variation ranged from 4.51% to 11.84%. Intraclass correlation coefficients ranged from 0.74 to 0.99. In healthy individuals, coefficients of variation ranged from 2.92% to 6.99%. Intraclass correlation coefficients ranged from 0.89 to 0.98. Intersession variability: In patients with glaucoma, coefficients of variation ranged from 3.68% to 10.50%. Intraclass correlation coefficients ranged from 0.82 to 0.99. In healthy individuals, coefficients of variation ranged from 3.13% to 6.92%. Intraclass correlation coefficients ranged from 0.87 to 0.99. Interexaminer variability: In patients with glaucoma, coefficients of variation ranged from 2.62% to 14.94%. Intraclass correlation coefficients ranged from 0.55 to 0.98. In healthy individuals, coefficients of variation ranged from 2.04% to 7.31%. Intraclass correlation coefficients ranged from 0.86 to 0.98.

Conclusions: These findings indicate that RNFL measurements with spectral-domain OCT display excellent reproducibility, with low intrasession, intersession, and interexaminer variabilities.