Accuracy of RTVue optical coherence tomography, Pentacam, and ultrasonic pachymetry for the measurement of central corneal thickness.

OBJECTIVE To compare the reliability and the agreement in measuring central corneal thickness (CCT) using the following technologies: RTVue Fourier-domain optical coherence tomography (Optovue, Inc., Fremont, CA), Pentacam (Oculus, Inc., Wetzlar, Germany), and ultrasonic pachymetry (USP; Pocket-II; Quantel Medical, Inc., Bozeman, MT). DESIGN Evaluation of diagnostic test. PARTICIPANTS One hundred four eyes of 52 healthy subjects (mean age ± standard deviation, 28.6 ± 4.8 years). METHODS One eye from each subject was assigned randomly for a repeatability test in which a single operator performed 3 successive measurements. The other eye underwent an interoperator reproducibility test by 3 operators. Two centering methods of RTVue and 3 types of CCT from Pentacam were investigated. For USP, 1 drop of topical anesthetic was administered, and measurement was initiated 90 seconds later. Agreement among the instruments was evaluated using Bland-Altman plots. MAIN OUTCOME MEASURES Various types of CCT were compared: central zone average and minimum thickness of RTVue centering on the vertex and the pupil; corneal thickness at the pupil center, apex, and thinnest location from Pentacam; and mean CCT of 5 repeated measurements of USP. The reliability of measurement was assessed using the repeatability or reproducibility coefficient (Rco), the coefficient of variation, and the intraclass correlation coefficient. The limit of agreement was used to analyze concordance. RESULTS The Rco of RTVue was 4 to 5 μm, which was comparable with that of USP and better than that of Pentacam (10-11 μm). The Rco was not dependent on centering methods (RTVue) or types of CCT (Pentacam). The location of minimum thickness found by RTVue was less reliable than that of the Pentacam. The central zone average of RTVue was approximately 7 μm larger than the pupil center or apex thickness of Pentacam and approximately 13 μm larger than the CCT measurement of USP. Those discrepancies could be as high as 20 and 23 μm, respectively. The minimum thickness measured by the RTVue was similar to that of Pentacam. CONCLUSIONS The RTVue is a rapid and reliable noncontact means of measuring CCT; however, the characteristics of CCT measured by RTVue must be understood when comparing the CCT obtained by the Pentacam or USP. FINANCIAL DISCLOSURE(S) The author(s) have no proprietary or commercial interest in any materials discussed in this article.