Further analysis of the size and shape of cells obtained by impression cytology from the exposed portion of the human bulbar conjunctiva.

PURPOSE

To assess whether the size of the cells obtained by conjunctival impression cytology can be quantitatively assessed by measurement of the longest dimension of the cells.

METHODS

Under topical benoxinate anaesthesia, cells were removed from the normally exposed nasal bulbar conjunctival surface using a 0.4 micron pore diameter filter (Biopore filter; type Millcell-CM). The filters were stained with haematoxylin after ethanol denaturation, photographed at 40 x magnification, and 35 mm slides prepared. An optical overlay method was used to outline the borders of sets of 30-35 contiguous cells on each image. The cell area, longest and shortest dimensions were measured by planimetry to an accuracy of +/- 3%.

RESULTS

Analyses of 20 sets of samples, from individuals aged 21 to 48 years and without clinically significant ocular surface disease, revealed a median cell area of 133 micron 2 (n = 621, range 46-1602 micron 2; average 212 micron 2), a median longest dimension of 13.9 microns (range 6.6-68.8 microns; average 16.9 microns) and a shorter dimension of 10.0 microns (range 4.7-43.0 microns; average 11.9 microns); the distributions of values indicated bimodality. Most cells had a long:short ratio (L:S ratio) value between 1.00 and 1.80, but 11.9 +/- 6.1% of the cells had L:S ratios between 1.80 and 4.60. The overall relationship between the longest dimension and the area of the cells was nonlinear, with cells having larger L:S ratios having disproportionately smaller areas.

CONCLUSIONS

The superficial conjunctival cells are small, and their longest dimensions are systematically related to area. Analyses of cell shape indicate further possible ways of identifying different cells on the bulbar conjunctiva. Compared to literature values, there is a substantial overlap in longest dimensions of the conjunctival cells with those of cells that can be collected off the corneal surface. This means that superficial conjunctival and corneal cells cannot be distinguished simply on the basis of measurements of the long dimension.