Liquid-based preparations versus conventional cytology: specimen adequacy and diagnostic agreement in oral lesions.

OBJECTIVE: To compare specimen adequacy and diagnostic agreement between liquid-based preparations and conventional smears in oral lesions, and to test the viability of immunocytochemical assay in liquid-based preparations from oral carcinoma lesions.

MATERIAL AND METHODS: Samples were collected from 44 patients. Conventional smears were prepared first, using a cytobrush device. Then the brush, containing the residual material, was immersed in a preservative fluid. The sample in the preservative fluid was processed according to the manufacturer directions (AutoCyte, Inc. Elon College, North Carolina, USA). Slides of both techniques were stained by Papanicolaou method. For immunocytochemical assay, a cytokeratin pool AE1/AE3 (Dako, CA, USA) was applied in liquid-based preparations from oral carcinoma lesions following the Streptavidin-biotin-peroxidase method. Fisher's exact test was used; significance was set for P < or = 0.05.

RESULTS: Both techniques agreed on cytologic diagnosis in every case they yielded an adequate specimen; in 3 cases conventional smear resulted in hypocellularity and therefore inadequate for analysis. On specimen analysis, the liquid-based cytology demonstrated a statistically significant, 41% overall improvement in smear thickness and 66% in cell distribution (P < or = 0.05), and a reduction in cell overlapping and presence of blood (P < or = 0.05). The cell morphology was better visualized in the liquid-based preparations. The immunocytochemical assay reactions were positive in all malignant cases, the visualization of the immunostained cells being especially clear.

CONCLUSION: Both, the liquid-based preparation and conventional smear, are diagnostically reliable; the liquid-based method showed an overall improvement on sample preservation, specimen adequacy, visualization of cell morphology and reproducibility.