[Data quality of unsupervised endothelial cell counting vs. reading centre analysis in multicentric clinical trials].
ABSTRACT The aim of this study was to assess data quality from unsupervised endothelial cell counting in the multicentric setting.
We performed an endothelial cell counting trial with two fictitious trial sites. The trial protocol simply demanded for marking 30 cells for analysis. Analyses were performed with the cell counting tool as built into the Topcon SP-3000P specular microscope. The first centre consequently dotted 30 cells. The other centre continuously dotted more cells until 30 cells were included in the cell counting analysis. Both sites analysed the same 89 eyes of corneal outpatients and heathy volunteers. Both sites used a dedicated Topcon SP-3000P microscope. The image pairs from both sites were eventually printed, scanned and re-evaluated with a programme that evaluated all visible cells ("reading centre"). The agreement between both sites was statistically assessed by means of Pearson's correlation and Bland-Altman analysis. The same statistical assessments were also performed for the image pairs as analysed in the reading centre.
The determined cell densities as reported by both trial sites differed by -65 % to 42 %. Furthermore, we also observed a systematic deviation between both sites. Consequently, the coefficient of determination from Pearson's correlation was only 0.947. However, the agreement was as high as 0.997 when the image pairs were analysed in the reading centre. Here the difference between the cell densities of the image pairs ranged from merely -15 % to 9 % with no systematic deviation.
Unsupervised endothelial cell counting does not result in sufficiently objective endothelial cell denstiy estimations. Furthermore, the built-in analysis tools can introduce systematic errors. Both drawbacks can be overcome by a reading centre that evaluates all visible cells on the images. For this reason, we recommend the involvement of a reading centre in multicentric clinical trials on the corneal endothelium.
- SourceAvailable from: plosone.org[Show abstract] [Hide abstract]
ABSTRACT: Modern specular microscopes (SM) robustly depict the same central area of the corneal endothelium at different time points through a built-in fixation light. However, repeated image acquisitions slightly shift and rotate because of minute changes in head position in the chin and forehead rest. This prevents the manual retrieval of individual corneal endothelial cells (CECs) in repeated measurements because SM images usually lack obvious landmarks. We devised and validated an image registration algorithm that aligns SM images from the same eye to make corresponding CECs coincide. We retrospectively selected 27 image pairs for the presence of significant image overlap. Each image pair had been recorded on the same day and of the same eye. We applied our registration method in each image pair. Two observers independently validated, by means of alternation flicker, that the image pairs had been correctly aligned. We also repeatedly applied our registration method on unrelated image pairs by randomly drawing images and making certain that the images did not originate from the same eye. This was done to assess the specifity of our method. All automated registrations of the same-day and same-eye image pairs were accurate. However, one single image incorrectly failed to trigger the non-match diagnosis twice in 81 registration attempts between unrelated images. As it turned out, this particular image depicted only 73 CECs. The average number of CECs was 253 (range 73-393). Repeated non-contact SM images can be automatedly aligned so that the corresponding CECs coincide. Any successful alignment can be considered as proof of the retrieval of identical CECs as soon as at least 100 CEC centroids have been identified. We believe our method is the first to robustly confirm endothelial stability in individual eyes.PLoS ONE 03/2013; 8(3):e59261. · 3.53 Impact Factor