Abstract
Background: Nuclear morphometry can be expected to improve the distinction between benign and malignant lesions. Patients and Methods: Forty fine-needle aspiration biopsy (FNAB) samples fixed in 50% ethanol were collected at the African Oncology Institute, Sabratha, Libya, during the period 2004-2007. All diagnoses reported were confirmed histologically. There were 23 cases of infiltrating ductal carcinoma and 17 of benign breast disease. Two different assessment methods were applied: measurements made on cell groups, and those made on free cells. Apocrine metaplasia was excluded. Five different size parameters (include mean nuclear area, MNA) and 6 shape factors were measured. Results: The size parameters showed significant differences between benign and malignant cases. The mean, median and 95% percentiles of nuclear area in both types of assessment were especially significant. The shape parameters were not significant. Conclusion: The study suggests that interactive computerized nuclear morphometry is an efficient and successful tool in distinguishing between cases of benign and malignant disease. Combination of our data with earlier free cell data gave the following diagnostic guidelines: Range of overlap in free cell samples: MNA 55 mu m(2) - 71 mu m(2). Cut-off values for diagnostic purposes: 100% detection of malignant cases: MNA>54 mu m(2) (specificity 84%), 100% detection of benign cases: MNA<72 mu m(2) (sensitivity 91%).