Abstract

DO.14.05

Evaluation of devices for cell counting in the endothelium and pachymetry of the cornea

Saadettin Sel1, K. Jendrischik1, Adriana Paraforos1, Anna Sasse1, C. Wicher1, D. Kaiser1, Thomas Kalinski2, Gernot I. W. Duncker1, Friedrich Paulsen3
1Universitäts-Augenklinik, Martin-Luther-Universität Halle-Wittenberg, Halle/Saale, 2Institut für Pathologie, Universität Magdeburg, 3Institut für Anatomie und Zellbiologie, Martin-Luther-Universität Halle-Wittenberg, Halle/Saale

Objective
The quality of the optical imaging of the eye is primarily determined by the functionality of the cornea. For the clinical application of newly developed instrumentation a reliability test of these devices to a reference device is necessary.
Methods
In this study, measurements for endothelium specifically and pachymetric data of 283 eyes are taken. Four different groups are defined for the survey. Two devices (Geraet 1 and Geraet 2) were compared with the reference devices NonCon Robo SP 6000/Konan (endothelium microscopy) and Pentacam/Oculus (pachymetry).
Results
The evaluation of the test results shows that three of the 32 sample comparisons show significant correlations of the measured values with the reference equipment. These are the comparison sample NonCon Robo SP 6000/Konan – Geraet 1 in the case group of eyes after keratoplasty regarding the percentage of hexgonal cells in the endothelium (p = 0148), the pair comparisons Pentacam/Oculus – Geraet 1 in the same case group (p = 0315) and the group of contact lens wearers (p = 0068) in relation to the corneal thickness. For all other comparisons no statistically relevant match is verifiable.
Conclusions
Under the assumption that the endothelium specifically data taken with the device NonCon Robo SP 6000/Konan and as well as the pachymetric measurements of the Pentacam/Oculus show the real actual anatomical-morphological characteristics, which also means that these both devices are considered to be the reference, the two devices (Geraet 1 and Geraet 2) supply statistically significantly deviating values.

Copyright: 2009 Porstmann Kongresse GmbH. All rights reserved. Impressum, Disclaimer.