To compare the results of specular microscopy with those of confocal microscopy for evaluation of corneal endothelium.
MethodsWe evaluated corneal endothelium of 103 eyes using specular microscopy and confocal microscopy. Endothelial cell density, pleomorphism, and polymegathism were measured using a ConfoScan 4 confocal microscope (Nidek Technologies, Inc, Greensborom, NC) in automatic mode before and after manual correction. Also, endothelial cell density, the coefficient of variation, and hexagonality were evaluated using a Konan Noncon Robo-8400 noncontact specular microscope (Konan medical, Inc., Hyogo, Japan). The differences in results obtained from these various methods were compared: polymegathism was compared with the coefficient of variation, and pleomorphism was compared with the inversion of hexagonality.
ResultsEndothelial cell density as measured by specular microscopy, the automatic count of confocal microscopy, and the manual correction for confocal microscopy were 2797.6±354.14 cell/mm2, 2973.1±284.24 cell/mm2, and 2861.9±335.58 cell/mm2, respectively. Results of each test was not significantly different ( p =0.241). The inversion of hexagonality, pleomorphism of automatically counted confocal microscopy, and the pleomorphism of manually corrected confocal microscopy were 56.14%, 54.77%, and 55.24%, respectively. Results of each test were not significantly different ( p =0.147).The coefficient of variation of specular microscopy, the polymegathism of automatic counted confocal microscopy, and the polymegathism of manually corrected confocal microscopy were 33.71%, 39.68%, and 38.75%, respectively. Results of each test were significantly different ( p =0.005).
ConclusionsEndothelial cell density and polymegathism as measured by confocal microscopy were not different from specular microscopy results in normal corneas, but these results were different for polymegathism in normal corneas. Therefore, manual correction for endothelial cell evaluation of a disordered cornea should be performed during clinical evaluation.