Purchase this article with an account.
Nimesh Bhikhu Patel, Kelsey Evans, Ronald S Harwerth; RNFL thickness measures using elliptical scans centered on the optic nerve head. Invest. Ophthalmol. Vis. Sci. 2014;55(13):4027.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The circumpapillary retinal nerve fiber layer (RNFL) thickness, quantified using spectral domain optical coherence tomography (SD-OCT), has become a clinical standard for the diagnosis and assessment of progression of optic neuropathy. Although it is not considered in clinical instruments, magnification effects should be included because RNFL thickness varies with distance from the optic nerve head (ONH) rim margin, the RNFL thickness depends on the location of the scan path, but the axonal content within this region should not. The purpose of this study was to compare an elliptical scan path with fixed distance from the rim margin to one with a constant circumference.
Raster, radial and circular SD-OCT scans, centered on the optic nerve, were acquired from 96 human subjects without history of ocular pathology. Transverse scaling for each eye was determined from ocular biometry using a schematic eye. Using Bruch’s membrane opening (BMO) as a reference, elliptical scans 900 µm from the rim margin and 10.7 mm in circumference, were interpolated from the ONH centered raster scan. Segmentation errors in the elliptical and circular B-scans were manually corrected. RNFL area was calculated by multiplying the RNFL thickness for each A-scan by its calculated width. The relationship between RNFL thickness/area and axial length was used to compare the accuracy of the measurements.
The mean RNFL thickness for standard 12 deg scan was 111.59±9.4 µm, 120.03±8.8 µm for elliptical scans 900 µm from the BMO and 117.50±8.0 µm for elliptical scans with a circumference of 10.7 mm. The mean RNFL area was 1.26±0.09 mm2, and not different across the scan locations (p=0.91). BMO area was larger in longer eyes (R2 = 0.21, p < 0.001) and global RNFL thickness for standard circular scans (slope = -3.4µm/mm, R2 = 0.27, p < 0.001), and elliptical scans 900 µm from the BMO (slope = -1.75µm/mm, R2 = 0.07, p = 0.004) decreased with increase in axial length. However, global thickness from elliptical scans with a 10.7 mm circumference were independent of axial length (p = 0.91).
An accurate assessment of the RNFL requires inclusion of individualized transverse scaling. Only scans with a fixed circumference demonstrate an independence from variation in axial length and should, therefore, provide the best surrogate for axonal content.
This PDF is available to Subscribers Only