Abstract
Abstract: :
Purpose: to determine how defined shifts in horizontal and vertical eye position affect the profile of RNFL thickness in order to understand the influence of eye movement. Methods: 7 normal subjects and 3 patients with glaucoma who demonstrated steady fixation had their RNFL thickness measured by OCT Stratus (fast RNFL circular scan=3 individual sequential scans). The circular scan was shifted from the centered location to 2 equally spaced locations (1/4 disc diameter and 1/2 diameter shifts) in each direction along the horizontal and vertical meridian (total of 5 scan locations/meridian). Results: When the scan was moved along the horizontal meridian, the superior and inferior RNFL humps were displaced further apart (circular scan moved toward temporal retina) or closer together (circular scan moved toward nasal retina), and the temporal and nasal thickness became asymmetric in thickness, inverse to one another. The RNFL thickness changed the most at the locations along the greatest slope of the RNFL TSNIT profile (maximal change = 50 microns for 1/4 disc diameter shift in scan position). A vertical shift in the circular scan caused the height of the superior and inferior arcuate RNFL thickness to become asymmetric; the shift was inverse and depended on the direction of the vertical scan movement. Glaucoma patients showed the same pattern of changes in RNFL thickness with scan position, but patients with severe thinning and flattening of the TSNIT profile showed less change from retinal position change relative to the scan. Conclusions: Eye position shift during a circular scan can be a major source of measurement variability and will affect the RNFL thickness TSNIT profile in a predictable way, depending on the direction and magnitude of eye position shift. The RNFL locations most susceptible are those on the greatest slope of the TSNIT profile with horizontal movements, and at the peak of the superior and inferior bundles with vertical movements. Position change in eyes with more severely thinned RNFL will have less influence on thickness measurement, but resolving smaller changes are required for detection of progression in these eyes. Interpretation of clock–hour abnormalities in the probability plot of the RNFL should take into account potential shifts in position of the retina with respect to the scan position or inaccurate scan position placement.
Keywords: nerve fiber layer • imaging/image analysis: clinical • imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound)