This cross-sectional, observational study was conducted with the approval of the Institutional Ethics Committee (Date: 17.10.2017/ No: 18/04). The study protocol followed the tenets of the Declaration of Helsinki. Informed consent was obtained from all the study participants.
The study population was selected from consecutive patients with keratoconus (keratoconus group) and subjects with minor refractive errors (control group) who presented at the Department of Ophthalmology Outpatient Clinic of Kırıkkale University between November 2017 and October 2018 and met the study criteria.
The diagnosis of keratoconus was made by evaluating the clinical and topographic features and accepted as the presence of at least one of the clinical findings of irregular scissoring reflex, Vogt stria, Fleischer ring, Munson sign, Rizutti sign, apical thinning, and an apical scar on retinoscopy and asymmetrical bowtie pattern suggesting keratoconus on corneal topography, paracentral steepening, and inferior superior asymmetry [7, 8].7,8 Keratoconus severity was evaluated according to the Amsler-Krumeich classification [9]. Stage-1 eyes were grouped as mild keratoconus, stage-2 eyes as moderate keratoconus and stage-3 and stage-4 eyes as advanced keratoconus.
The study exclusion criteria were defined as intense corneal scarring, inadequate capture quality, history of corneal cross-linking or surgery in the keratoconus group; and for the control group, a refractive error of > ± 2 D of spherical equivalent or > ± 1 D of astigmatism, irregular astigmatism, confirmed or suspected keratoconus, best-corrected visual acuity (BCVA) < 20/20 (Snellen), intraocular pressure (IOP) > 21 mmHg, or a history of ocular surgery. Subjects wearing rigid gas-permeable or soft contact lenses were excluded from the study in both groups. If a patient had manifest keratoconus in only one eye, that eye was included in the study, whereas in patients with bilateral manifest keratoconus a randomly selected single eye was included in the analysis, as was the case for the control group subjects.
Measurements
All the study subjects underwent a routine ophthalmology examination, including visual acuity assessment, refractive error, slit-lamp biomicroscopy, IOP (after the anterior segment measurements), and fundus examination.
Anterior segment measurements were made on the same day using AL − Scan OB and Sirius CT devices in random order. All of the measurements were repeated 3 times consecutively by the same observer (YA) on each device in the same session. The patient was asked to focus on the central fixation light during the measurement, and blink for 10 seconds between the captures. Three consecutive captures, which were specified as adequate, were included in the study. Central corneal thickness (CCT), anterior chamber depth (ACD), white-to-white distance (WTW), and keratometry (K) data were recorded on both devices.
Sirius CT measures up to 151200 points on the front surface and up to 145600 points on the rear surface, through a topographic image of 22 Placido disc rings and cross-sectional 25 meridional Scheimpflug images of the anterior segment. Anterior corneal surface information is gathered by combining the data of the Placido and Scheimpflug systems, whereas data on the posterior surface of the cornea, the anterior surface of the lens, and the iris are obtained using solely the Scheimpflug camera. Corneal thickness, tangential and axial curvatures of the anterior and posterior corneal surfaces, the refractive power of the cornea, and corneal wavefront maps are produced by the software [10]. The default output provides keratometry data of Sim –K (average curvature value between 4th and 8th Placido rings), and curvatures at the 3.0 mm, 5.0 mm, and 7.00 mm zones, of which the first two were included in this study.
AL − Scan OB uses partial coherence superposition of light waves emitted from a superluminescent diode laser to measure biometric features of the eye such as axial length, corneal diameter, and keratometry and the Scheimpflug principle to measure CCT and ACD [11]. A ring image is projected on the patient’s cornea and a photodetector calculates the keratometry (corneal curvature radius) values and the angle of the flattest and the steepest meridians are calculated over 2.4 and 3.3 mm diameters[5].
Statistical Analysis
Sample size was calculated using G power software (version 3.1.9, Heinrich-Heine-Universitat Düsseldorf). All data were analyzed statistically using SPSS 21.0 software (IBM Corpn.) and MedCalc (version 11.6.0.0; MedCalc Software bvba). Conformity of the data to normal distribution was examined using the Kolmogorov–Smirnov test. Descriptive statistics were presented as mean ± standard deviation values. The Chi-square test was applied to comparisons of categorical variables. The first of the 3 measurements was taken into account for device comparisons, and a paired t-test or Wilcoxon signed ranks test was used. A value of p < 0.05 was considered statistically significant.
“Intra-session repeatability” refers to the variability between repeated measurements made on the same subject by the same observer in the same session. In this study the intra-session repeatability was evaluated with the following 4 parameters: 1) within-subject standard deviation (Sw): the standard deviation of 3 consecutive measurements taken from the same eye, 2) intrasubject precision (IP): the difference between the measured value and the actual. It is expected that 95% of the observations will be below the number calculated by the formula [1.96 x Sw]. 3) Repeatability index (RI): also known as test-retest repeatability. It is calculated with the formula [2.77 x Sw] under the assumption that the errors are normally distributed. RI shows the size of the error range based on the difference between the measurements repeated with the same method in 95% of the observations. The smaller the RI, the better the repeatability. To say that the measurement is repeatable, this index must be less than the mean of the studied parameter. 4) Coefficient of variability (CV): It is calculated with the formula [100 x (Sw/Mean of 3 measurements taken from the same eye)]. The CV value of a parameter is calculated by averaging the CV values in all eyes and is expressed as a percentage. The lower this coefficient, the lower the variation of the measurement and the better the repeatability [12].
Intra-observer reliability means obtaining the same result in the same subject under the same conditions, with the same observer and instrument. Reliability is expressed by the intraclass correlation coefficient (ICC). The ICC reflects the consistency of the datasets of repeated measurements and ranges from 0 to1. A CCT coefficient < 0.50 indicates low reliability, 0.50–0.75 moderate reliability, 0.75–0.90 good and > 0.90 excellent reliability [13]. In this study, the ICC value of 3 measurements for each parameter was calculated using a two-way mixed-effect model with the absolute agreement method.
The Bland–Altman plots were used to assess the agreement between the devices. The x-axis in the graphs displays the mean value of the two instruments, and the y-axis shows the difference. The 95% limits of agreements (LoA) are defined as the mean difference ± 1.96 SD between the AL − Scan and the Sirius [10].