Central corneal thickness measurements obtained with anterior segment spectral domain optical coherence tomography compared to ultrasound pachymetry in healthy subjects.

Introduction: Central corneal thickness (CCT) imparts information about an individual's risk of conversion to glaucoma from ocular hypertension, progression of established glaucoma, and the likelihood of developing structural and functional abnormalities in patients with ocular hypertension. Most typically, CCT is measured through ultrasound (US) pachymetry. Currently, optical coherence tomography (OCT) has the ability to image the anterior segment, cornea, and anterior chamber angle. With this ability comes the option of determining CCT. The purpose of this study is to ascertain any significant difference in CCT measurement results as well as quantify the reproducibility of measurements of the two technologies. In addition, by measuring CCT both with traditional US pachymetry as well as spectral domain (SD) OCT technology, we sought to determine if CCT measurement by SD-OCT is an accurate, comparable and viable option;

Methods: Eighty eyes of forty healthy volunteers were used to determine CCT with both SD-OCT and US pachymetry. Three consecutive measurements were collected with each method on every eye.

Results: CCT measurements made by US pachymetry and SD-OCT were similar and consistent (r=0.99 for both methods). CCT measurements made by SD-OCT were consistently thinner by approximately 12 micrometers than measurements made by US pachymetry. Repeated measurements of CCT obtained by SD-OCT were more reproducible and had less variability than measurements obtained by US pachymetry. The mean within-subject standard deviation among SDOCT was significantly smaller than that in US pachymetry (1.92 in SD-OCT vs. 2.04 in US pachymetry, p=0.036);

Conclusions: Measurement of CCT by SD-OCT compares favorably with and is at least as accurate as measurements made by US pachymetry. Repeat measurements of CCT by SD-OCT have less variability than those obtained by US pachymetry, are more reproducible, possibly more reliable, and may better represent actual CCT.