Short-term repeatability of hand-held keratometry measurements

The Collaborative Longitudinal Evaluation of Ethnicity and Refractive Error (CLEERE) Study is a multi-center, observational investigation of ocular component and refractive error development in schoolchildren. Anterior corneal curvature is one of several ocular components which influence refractive status of the eye, and the CLEERE Study uses the Alcon Auto-Keratometer to measure corneal curvature. This report assesses the repeatability of this hand-held instrument. Previous studies have demonstrated the validity of the Alcon Auto-Keratometer; however, none have assessed its repeatability. Sixty children were recruited from clinics affiliated with the Southern California College of Optometry and the University of Houston College of Optometry. Two sets of five hand-held keratometry measurements were obtained on the right eye of each child by one investigator at each site using the Alcon Auto-Keratometer. The two sets of measurements were performed at least 10 minutes apart. The mean differences between the two occasions were not significantly different from zero for either the flat or steep corneal curvature measurements regardless of the number of readings taken. The largest improvement in repeatability, primarily for the steep meridian, occurred when the first two readings were averaged. The addition of readings 3, 4, and 5 to the average did not substantially improve repeatability for either meridian. The 95% limits of agreement between the average of two readings on two occasions for the flat and steep meridians were +/-0.28 and +/-0.39 D, respectively. The 95% limits of agreement after two readings were +/-0.28, +/-0.20, and +/-0.24 D for the M, J0, and J45 vectors, respectively. The Alcon hand-held keratometer provides a repeatable measure of corneal curvature as demonstrated by short-term repeat agreement within +/-0.50 D. This level of repeatability can be achieved only by manually averaging two consecutive measurements.