applied to compare shape and size variations of the basal nuclei of the laryngeal epithelium in normal, laryngeal intraepithelium neoplasia (LIN) and invasive carcinoma to assess the reliability of light microscopic criteria used in grading dysplasia according to Friedman classification. Morphometrical analysis was carried out by Shape Analytical Morphometry (S.A.M) system. The logical architecture assumes that each irregular shape contains elements of two distinct logical domains: gross distortions that interest the contour and its local perturbations. These features were investigated separately by analytical procedures to acquire independent parameters both on the logical level and the numerical one. The nuclear area significantly increased from normal to carcinoma (p<.001). The increasing of the nuclear area was evident also in LIN I. Nuclear distortions were present in LIN II and LIN III. The highest nuclear contour irregularities were found in LIN III. Multivariate analysis showed a difficulty in discriminating various grades of dysplasia, especially between LIN I and LIN II (31% of error). In conclusion, our results indicate that nuclear pleomorphism of the basal cells layer, using a unique evaluator, is an unsatisfying criterion to distinguish moderate dysplasia.
Quantitative morphometrical investigation of basal cell layer in laryngeal premalignant lesions
SERIO, Gabriella;GIARDINA, Carmela;Lettini T;RESTA, Leonardo
2002-01-01
Abstract
applied to compare shape and size variations of the basal nuclei of the laryngeal epithelium in normal, laryngeal intraepithelium neoplasia (LIN) and invasive carcinoma to assess the reliability of light microscopic criteria used in grading dysplasia according to Friedman classification. Morphometrical analysis was carried out by Shape Analytical Morphometry (S.A.M) system. The logical architecture assumes that each irregular shape contains elements of two distinct logical domains: gross distortions that interest the contour and its local perturbations. These features were investigated separately by analytical procedures to acquire independent parameters both on the logical level and the numerical one. The nuclear area significantly increased from normal to carcinoma (p<.001). The increasing of the nuclear area was evident also in LIN I. Nuclear distortions were present in LIN II and LIN III. The highest nuclear contour irregularities were found in LIN III. Multivariate analysis showed a difficulty in discriminating various grades of dysplasia, especially between LIN I and LIN II (31% of error). In conclusion, our results indicate that nuclear pleomorphism of the basal cells layer, using a unique evaluator, is an unsatisfying criterion to distinguish moderate dysplasia.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.