Intra- and interobserver reproducibility of Bruch’s membrane opening minimum rim width and retinal nerve fiber layer thickness in healthy individuals with spectral domain optical coherence tomography

Abstract

Purpose : To investigate the intra- and interobserver reproducibility of Bruch’s membrane opening minimum rim width (BMO-MRW) and peripapillary retinal nerve fiber layer (RNFL) thickness measurements in a population of healthy individuals with spectral domain optical coherence tomography (SD-OCT).

Methods : Fifteen healthy subjects were included in the study. Subjects had normal eye examination and normal visual fields. One eye of each subject was randomly selected as the study eye. Optic nerve head (24 radial scans centered on BMO) and peripapillary circle scans (3.5 mm nominal scan diameter) were acquired relative to the fovea/BMO center axis (FoBMO) using SD-OCT (Spectralis, Heidelberg) by three examiners. Additionally, one examiner imaged each patient three times. Intra- and interobserver reproducibility of global BMO-MRW and RNFL thickness measurements were determined by coefficient of variance (COV) and intraclass correlation coefficient (ICC).

Results : The median (interquartile range, IQR) age was 34.9 (26.6, 41.8) years. The median (IQR) global BMO-MRW was 333.2 (301.5, 369.5) μm and the corresponding values for RNFL thickness were 101.0 (96.0, 108.0) μm. The intraobserver within-subject standard deviation (SD) and 95% confidence interval (CI) for BMO-MRW and RNFL thickness were 2.7 (2.0, 3.4) μm and 1.2 (0.6, 1.9) μm, respectively. The corresponding interobserver values were 4.0 (3.0, 5.0) μm and 2.4 (1.8, 3.0) μm. The intraobserver ICC (95% CI) for BMO-MRW and RNFL thickness were 0.996 (0.991, 0.998) and 0.968 (0.931, 0.988), respectively. The corresponding interobserver values were 0.992 (0.982, 0.997) and 0.905 (0.799, 0.962). The intraobserver COV for BMO-MRW and RNFL thickness were 0.9% and 1.2%, respectively. The corresponding interobserver values were 1.3% and 2.1%. The reproducibility values are summarized in Table 1.

Conclusions : Spectralis OCT shows excellent reproducibility for measuring both BMO-MRW and RNFL thickness in healthy subjects, with a tendency of higher reproducibility with BMO-MRW.

This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.

 

Table 1. Intra- and interobserver values of Bruch’s membrane opening minimum rim width (BMO-MRW) and peripapillary retinal nerve fiber layer (RNFL) thickness. SD, standard deviation. CI, confidence interval. ICC, intraclass correlation coefficient. COV, coefficient of variance.