Abstract
Abstract: :
Purpose: To investigate the effects of refractive error (RE) and axial length (AL) on quantitative optical coherence tomography (OCT) measurements. Methods: Scan lengths on the retina were calculated using OCT (Zeiss-Humphrey Systems Inc., Dublin, CA) by inputting various RE and AL values. Peripapillary circular OCT scans were performed on eyes with glaucoma (mean deviation ≷-5.0 dB) to measure the retinal nerve fiber layer (RNFL) thickness. RE was entered for all eyes. On-screen circular scan length reading was maintained to 10.86 mm. AL was first set to the default value (24.46 mm) and then the actual value obtained using A-scan ultrasound. Results: The Humphrey software assumes a default setting of 0.0 diopters RE and AL of 24.46 mm. By altering the AL and RE, both individually and in combination, the actual scan length (SL (mm)) and sampling interval (SI (mm); distance interval between adjacent sampling points projected on retina) was affected (see table). With a refractive error range of -10 to +10 diopters the change in scan length was ± 2%, while with an AL range of 18.0 to 30.0 mm the change of the scan length was -20% to +27% from the default AL. Thirteen subjects (mean age 51.2 ± 14.5 (SD) yrs., range 22 to 66 yrs.; mean axial length 25.03 ± 1.67 mm, range 22.86 to 28.51) were enrolled. There was no difference in mean or quadrant RNFL thickness measurement with or without AL correction (all p≷0.11, paired t-Test). The mean RNFL thickness difference was 4.2 ± 3.4 microns (range 0 to 10 microns), which is within the physical resolution range. Conclusion: RE information (±10 D) has a negligible impact on OCT measurements. AL information does affect sampling density and measured RNFL thickness. Table
Keywords: 430 imaging/image analysis: clinical • 432 imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • 356 clinical (human) or epidemiologic studies: systems/equipment/techniques