Abstract
Purpose :
The presence of retinal layers other than retinal nerve fibre layer (RNFL) within the optic nerve head (ONH) minimum rim width (MRW) was previously described in monkeys (Fortune B et al., IOVS 2016). We aimed to determine the occurrence and relevance of these layers within the MRW, herein referred to as “protruded layers” (PL), in normal human subjects.
Methods :
24 radial ONH B-scans, centred on the Bruch’s membrane opening and aligned to the fovea, were acquired from 30 healthy subjects with optical coherence tomography (Spectralis). One randomly selected eye per subject was analyzed. In each of the 2 positions for each B-scan, the MRW was examined for the presence of PL, which was manually measured when detected (Fig 1). The proportion of PL comprising the MRW was calculated. To evaluate the reproducibility of PL detection and measurement, 2 observers (Observers 1 and 2) analyzed a subsample of 240 scans from 15 subjects.
Results :
The median age of the subjects was 69 years (interquartile range [IQR]: 63.2 – 76.7 years). Of the 240 B-scans evaluated for reproducibility, Observer 1 detected PL in 36 (15%) positions and Observer 2 in 38 (16%) positions. The observers agreed about the presence of PL in 31 of 43 positions (72%) and about the PL’s absence in 188 of 208 positions (90%). The median PL thickness difference among the observers was 4.0 µm (IQR: -3.0 µm – 16.5 µm). Of all the 720 B-scans analyzed, PL was detected in 199 (14%) of all 1440 positions. Each eye had a median of 8 positions out of 48 (17%) with PL (IQR: 3.5 [7%] – 14.5 [30%]). 23 (77%) of the subjects had at least one position with PL. The temporal half of the ONH contained 183 (91%) of all 199 positions with PL (Fig. 2). When present, the median PL thickness was 65 µm (IQR: 48 µm – 86.5 µm), making up a median of 28% (IQR: 18% - 38%) of the MRW.
Conclusions :
The frequent presence of PL in the ONH radial scans, mainly in the temporal quadrant, and the variable proportion of PL that constitutes the neuroretinal rim, may impact diagnostic accuracy in single examinations. However, we expect the presence of PL to have minimal impact in detecting change over serial examinations.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.