To evaluate investigator- and instrument-dependent errors of keratometry in children and their possible influence on the predicted postoperative refraction after intraocular lens implantation.
Methods: Keratometry readings of five children with congenital cataract were taken preoperatively under general anesthesia (age 1.5-7.5 months, mean = 4.4 months). Both eyes were measured by two investigators with a manual Zeiss keratometer and an Alcon portable automated keratometer. Three measurements were performed with each instrument. Axial length was determined with ultrasound and the SRK II-formula was used for intraocular lens calculation.
Results: The investigator-dependent variability was 0.22 mm +/- 0.07 SEM for the manual keratometer and 0.20 mm +/- 0.07 SEM for the automated keratometer. The instrument-dependent variability was 0.44 mm +/- 0.12 SEM for investigator A and 0.34 mm +/- 0.09 SEM for investigator B. These results suggest that deviation from the required postoperative refraction of up to 6.0 D has to be expected in individual cases if intraocular lenses are implanted.
Conclusions: The lack of fixation in children who have keratometry under general anesthesia leads to inaccurate keratometry readings, which will cause rather high deviations of the precalculated postoperative refraction. Our results explain the high rate of residual refractive errors reported in the literature after intraocular lens implantation in children. In order to improve the accuracy of keratometry multiple measurements of corneal curvature should be taken.