Meta-analysis of Pentacam vs. ultrasound pachymetry in central corneal thickness measurement in normal, post–LASIK or PRK, and keratoconic or keratoconus-suspect eyes
AIM To evaluate the repeatability of central corneal thickness (CCT) measurement by entacam, and agreement of CCT measured by Pentacam and ultrasound pachymetry (USP) in Chinese myopia. Thereby investigate the possibility of Pentacam as a substitute for USP in CCT measurement before refractive surgery. The effects of corneal curvature measured by Pentacam on CCT were also evaluated. METHODS One hundred and forty-eight right eyes of 148 individual with myopia were included in this study. Three successive Pentacam CCT measurements followed by 10 successive ultrasound pachymetry were carried out in the 148 eyes. Mean of CCT taken by each device was calculated for comparison. According to the CCT measured by USP, all the 148 eyes were divided into 3 groups: <520µm, 520-560µm, >560µm. For all eyes and each group the CCT obtained by Pentacam and USP were compared. Anterior corneal curvature of the 148 eyes was also adopted for correlation analysis with CCT obtained by ultrasound pachymetry. In addition, CCT measurement using 60 random selected Scheimpflug images was performed by 3 skilled investigators at different time, and this was repeated for 3 times by a forth investigator to assess repeatability of Pentacam CCT measurement using Scheimpflug images. RESULTS Intraclass correlation coefficient (ICC) analysis revealed high intraobserver repeatability (ICC=0.994, F=158.60, P<0.001) for CCT measurement by Pentacam. The interobserver (ICC=0.998, F=494.73, P<0.001) and intraobserver (ICC=0.997, F=383.98, P<0.001) repeatability for Pentacam CCT measurements using Scheimpflug images were also excellent. There was high positive correlation between the CCT values measured by Pentacam and ultrasound pachymetry (r=0.963, P<0.001). Bland-altman plots showed that the Pentacam underestimate the CCT by 8.02µm compared with ultrasouond pachymetry. The differences between Pentacam and USP increased as the CCT readings by USP increased (Pentacam vs USP: slope=-0.04, P<0.05). The 95% upper and lower limits of agreement between CCT values obtained from the two devices were +9.33µm and -25.37µm. No significant association could be found between CCT and anterior corneal curvature. CONCLUSION Inter- and intraobserver variability for CCT measurements by Pentacam was considerably below clinically significant levels. CCT of myopia obtained by Scheimpflug camera, Pentacam, were highly correlated to that by ultrasound pachymetry. However, the values obtained are not directly interchangeable between Pentacam and ultrasound pachymetry as the 95% limits of agreement are relatively wide. Pentacam can be a useful instrument for measuring CCT in candidates to refractive surgery in clinic.