Background: It is well recognized that distinguishing benign from malignant papillary lesions of the breast may pose challenging diagnostic problems. To prospectively evaluate the potential role of mammography, ultrasound and image-guided core biopsy in the diagnosis of papillary lesions of the breast.
Methods: 1,442 women consecutively underwent 14-gauge core biopsy and in 51 cases (3.5%) a diagnosis of papillary lesion was formulated. Both radiologists and pathologists independently expressed their degree of suspicion of malignancy (not suspicious, low, moderate, high) on the basis of radiological and core biopsy findings, respectively. Surgical excision of the lesion was used as gold standard and diagnostic agreement was assessed by the kappa statistic.
Results: At surgery, 19 of the 49 (38.7%) resected cases had a diagnosis of malignancy. A poor agreement was found between mammography and core biopsy results in the categorization of suspicion of malignancy (k = 0.03). Similar data were obtained between ultrasound and core biopsy (k = 0.07). A poor agreement was also observed between radiological and surgical results (k < 0.20). In contrast, a good agreement was found between core biopsy and surgical samples (k > 0.70). However, 5 (26%) out of the 19 malignant cases at surgery were judged as benign or probably benign on core biopsy. Depending on how the categories of suspicion on core biopsy were set up, the range of sensitivity was 74-89%, whereas specificity ranged from 91 to 97%.
Conclusions: Image-guided large core biopsy allows for a correct diagnosis in the majority of papillary lesions. However, its sensitivity is not good enough for surgical excision to be avoided.
Copyright 2003 S. Karger AG, Basel