Repeatability and Reproducibility of Central Corneal Thickness Measurement With Pentacam, Orbscan, and Ultrasound

Abstract

The purpose of this study was to compare central corneal thickness (CCT) measurements obtained with a novel rotating Scheimpflug camera (Pentacam; Oculus) with scanning slit topography (Orbscan; Bausch & Lomb), and with ultrasound pachymetry (SP-2000; Tomey). CCT in 30 healthy eyes was measured twice with each modality by 2 independent observers in random order. The results from scanning slit topography are given both with and without multiplication with the "acoustic correction factor" of 0.92. In addition, the displayed images from the rotating Scheimpflug camera and scanning slit topography were used to calculate the signal difference-to-noise ratios (SD/N) between cornea and background signal. The mean CCT values as determined with the different modalities (+/-standard deviation) were: 542+/-29 microm, 576+/-37 microm, 530+/-34 microm, and 552+/-32 microm for rotating Scheimpflug imaging, for uncorrected and for corrected scanning slit pachymetry, and for ultrasound, respectively. The differences between modalities (+/-95% limits of agreement) were -9.8+/-31 microm between rotating Scheimpflug and ultrasound, 24+/-31.2 microm between scanning slit and ultrasound, and 33+/-27 microm between scanning slit and rotating Scheimpflug imaging. The limits of agreement for within and between observer effects were within 4.2% of the absolute CCT values for scanning slit and ultrasound and within 2.2% for the rotating Scheimpflug imaging. The rotating Scheimpflug camera showed similar SD/N ratios but steeper edges of the corneal surfaces in the intensity profile plots. In the assessment of normal corneas, the Pentacam measured CCT values closer to ultrasound pachymetry and with less variability compared with Orbscan. The (interobserver) reproducibility with the Pentacam was highest of all 3 modalities.