Abstract
Purpose :
Spectral domain optical coherence tomography (SD-OCT) is useful for assessing retinal pathology. One important measured parameter is the retinal thickness. On SD-OCT devices, a default cornea curvature is assumed when generating the retinal thickness. The effect of using individualised cornea curvature on retinal thickness measurements is unknown. This study aimed to evaluate the effect of cornea keratometry on retinal thickness measurements obtained from SD-OCT.
Methods :
In a prospective study, 64 healthy volunteers were recruited and underwent auto-keratometry, auto-refraction and SD-OCT scans using the Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany). Retinal thicknesses were evaluated with the default cornea curvature of 7.7mm, following which, the actual keratometry was used. Comparisons between original and adjusted thicknesses were made using paired t-tests.
Results :
The mean retinal thickness was 263.0 µm using the default cornea curvature, and 262.4 µm using the actual keratometry (difference of 0.55 µm, p=0.006). The differences in retinal thicknesses were significant in the outer retina subfields (p< 0.02) but not in the inner retina subfields. The difference in retinal thickness was greater for emmetropes followed by low myopes and high myopes (0.87 µm, 0.58 µm and 0.44 µm respectively). Eyes with keratometry <42D had a decrease in central retinal thickness (mean = 2.8 µm) whereas eyes with keratometry >44D had an increase in central retinal thickness (mean = 0.93 µm).
Conclusions :
Cornea curvature has a significant effect on the measured retinal thickness output on the Spectralis OCT. Using corneal keratometry measurements increases the accuracy of retinal thickness measurements.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.