Abstract
Purpose: :
To compare Keratometry (K) readings by six
Methods: :
Scanning-Slit Technology (SST), three Scheimpflug Imaging Systems (SIS), and Placido Disk/Optical coherence tomography (PD/OCT) and Automated Keratometry (AK)And also Corneal Anterior Elevation (CAE) and Corneal Posterior Elevation (CPE) measurements by 5 methods: SST,3 different SIS and PD/OCTMethods: This was a prospective single-center study, of 84 eyes of 42 patients were were examined with the the Orbscan II (SST) the Pentacam, Galilei, and Sirius systems (all three SIS), the Atlas/Visante (PD/OCT) and the IOL master (IOLM)To assess the intra/interobserver and interdevice reproducibility of the 6 systems, all the eyes were examined 2 times by 2 independent observers and the reliability coefficient was calculated using Cronbach alpha-test
Results: :
The mean flat, steep and average K were (diopters): 43.05 ± 2.45, 45.86 ± 3.65, 44.32 ± 2.81 D (Orbscan), 42.74 ± 2.32, 45.26 ± 3.84, 44.04 ± 2.71 D (Pentacam), 43.18 ± 2.27, 45.68 ± 3.04, 44.43 ± 2.56 D (Galilei), 42.86 ± 2.27, 45.42 ± 3.22, 44.10 ± 2.63 D (Sirius), 42.99 ± 2.28, 45.42 ± 3.40, 44.21 ± 2.76 D (Atlas/Visante), 43.08 ± 2.59, 45.86 ± 3.65, 44.47 ± 3.01 D (IOL Master)The mean CAE and PCE were (microns): 14 ± 14 μm , 37 ± 28.3 μm (Orbscan), 5.5 ± 5.5 μm, 11.87 ± 15.09 μm (Pentacam), 2.57 ± 2.83 μm, 5.33 ± 7.18 μm (Galilei), -1.46 ± 1.65 μm, -3.30 ± 3.96 μm (Sirius), 7.20 ± 4.87μm (Atlas/Visante)Intraobserver and interobserver Cronbach alpha-test showed a very good reliability coefficient (>0.90) in all measurements in all the systemsUsing the Pentacam as reference the Inter device reliability coefficient for K readings was very good for SST, two SIS and the PD/OCT (>0.9 0), but not for the IOL master (less than 0.50)For the Corneal elevation the Inter device reliability coefficient was good (>0.80)
Conclusions: :
K, CAE and CPE measurements with the SST, the three SIS systems, the IOL Master and the PD/OCT showed very good correlation and high reproducibility in the Inter and intra observer reliability coefficientsNevertheless differences between the interdevice K readings with the IOLM were significant and there was poor correlationAlthough there were good intra-inter observer correlation in the CAE and CPE the Standard Deviations were high specially in the CPE, this could be due to the inclusion of patients with keratoconus in the study, but this does not affect the main outcome of our study purposeThis Results show that the K, the CAE and CPE measurements with several topography systems can be used interchangeably, but the IOLM K readings may be not accurate enough for clinical screening, and surgical decision, specially for refractive treatments