Subjective vs. objective analysis of the corneal endothelial cells in the rabbit cornea by scanning electron microscopy - a comparison of two different methods of corneal fixation

OBJECTIVE: To illustrate a method of quantitative scanning electron microscopy (SEM) for the evaluation of the corneal endothelia of small animals by comparing two commonly used fixation methods. ANIMALS STUDIED: Female New Zealand white rabbits, aged 10-12 weeks. PROCEDURES: The corneas were either dissected from the eye and placed in fixative (2% glutaraldehyde in 80 mm cacodylate, pH 7.4) or the whole eyeball immersed in fixative solution and the cornea dissected later. Lower (x 200) and high magnification (x 1,000-5,000) images were inspected for overall appearance. Magnification images (x 500) were used to measure the areas of 100 cells, and cell density (CD) calculated from the average area. RESULTS: Both fixation protocols yielded an intact endothelial surface, but the dissect- then-fix protocol resulted in more creasing and distortion artifacts that were avoided with the whole-eyeball fixation. Overall, the CD values were higher if the dissect-then-fix method was used, and the uniformity of the cell mosaic was less. The median CD values for the central, mid-peripheral and peripheral regions following dissect-then-fix protocols were 7,693, 7,353, and 7,071 cells/mm(2) (average eight corneas at each location). If the whole-globe fixation was used, the median CD values were 6,098, 5,747, and 4,785 cells/mm(2). CONCLUSION: Cell density values in SEM can be very different according to the fixation method used. A distinct regional difference in CD was evident, which was more pronounced if the cornea was fixed prior to being dissected from the eye.