Consistency of corneal sublayer thickness measurements using Fourier-domain optical coherence tomography after phacoemulsification

To assess the reliability of corneal epithelial thickness (CET), nonepithelial central corneal thickness (NECCT), and central corneal thickness (CCT) measurements using Cirrus high-definition optical coherence tomography (HD-OCT) in patients who did and did not undergo cataract surgery.

Forty patients who underwent uneventful phacoemulsification and 40 healthy participants were recruited to evaluate the intraobserver repeatability and interobserver reproducibility of CET, NECCT, and CCT measurements using Cirrus HD-OCT. To analyze repeatability, one examiner obtained 5 consecutive scans in each participant; for interobserver reproducibility, another examiner randomly obtained another scan. Within-subject standard deviation, coefficient of variation (CV), limits of agreement, and intraclass correlation coefficient (ICC) data were obtained.

For intraobserver repeatability, the intrasession CV (CVw) and ICC values of the CET in the operated and nonoperated groups were 3.7% and 0.80 and 3.8% and 0.73, respectively; for NECCT, 0.7% and 0.98 and 0.8% and 0.97; and for CCT, 0.6% and 0.99 and 0.7% and 0.98. For interobserver reproducibility, the CVw and ICC values for the CET in the operated and nonoperated groups were 2.6% and 0.82 and 2.3% and 0.62, respectively; for NECCT, 0.7% and 0.98 and 0.5% and 0.98; and for CCT, 0.5% and 0.99 and 0.4% and 0.99.

The corneal sublayer thickness can be measured reliably using Cirrus HD-OCT in patients who underwent cataract surgery and elderly participants; however, the CET consistency is poorer than the NECCT. Corneal epithelial thickness modifications exceeding 4% reflect true thickness changes instead of random error variations using HD-OCT.