Comparison of morphology of bulbar conjunctival cells assessed by impression cytology versus scrape and smear methods

Purpose: To assess the morphology of superficial bulbar conjunctival cells collected by impression cytology or by scraping. Methods: Freshly procured sheep eyes from a slaughterhouse were washed and then conjunctival impression cytology (CIC) samples taken from the nasal aspect of the bulbar conjunctiva on a Biopore Millicell filter unit, or cells scraped from the same location and smeared onto glass slides. After alcohol fixation, all samples were stained with Giemsa. Cell images were obtained by light microscopy at 200× magnification, and subjected to morphometry. Results: Cells obtained by scrape/smear processing appeared larger and less densely staining than cells procured by CIC. This appearance was confirmed by morphometry with measures of cell longest dimension LONG being 27.7 ± 1.3 μm and 21.7 ± 1.5 μm, respectively, while cell AREAC measures averaged 543 ± 42 μm²and 333 ± 51 μm²; these differences were significant (p < 0.001). Similarly, cell nucleus length (NUCLONG) and nucleus area (AREANUC) were also larger in smeared cells (p < 0.001). Various calculations of nucleo-cytoplasmic ratios as based on dimensions or area were only slightly different when comparing cells obtained by the two methods, indicating that the glass slide-smeared cells simply spread out more than cells on the CIC filters. Conclusions: Differences in cell and nucleus size can result from use of different processing methods. Therefore, in making comparisons between sets of conjunctival cells, it is important that cell collection and processing methods be considered.