Histological correlation of mammographically detected microcalcifications in stereotactic core biopsies

The aims of this study were to assess the value of specimen radiographs of stereotactic core biopsy, the usefulness of measuring size of calcifications on tissue sections, whether demonstration of calcifications in tissue sections alters the pathological diagnosis when specimen radiograph demonstrates calcifications, and to correlate these assessments with diagnostic outcome. A total of 301 core biopsies from 266 women with 274 mammographically suspicious areas of calcifications were examined. Core biopsies (five cores per procedure) were obtained stereotactically using a 14-gauge needle in an automated Biopty gun. Prior to processing of the tissue, 214 core biopsy specimens from 193 women with 197 lesions were radiographed. Of the 301 core biopsies, 56 (19%) were diagnosed as malignant, 15 (5%) were diagnosed as atypical ductal hyperplasia and 230 (76%) contained benign breast tissue. Of the core biopsies diagnosed as benign, 160 (70%) had specimen radiography prior to processing. Of these, 109 (69%) core biopsies showed calcifications on specimen radiographs. In 96 (88%) of these core biopsies, calcifications measuring > 100 microm were found on the initial tissue sections. In 11 (10%) further deeper sections were required to detect calcifications > 100 microm; however, this did not result in a change of the pathological diagnosis. Two of the 109 (1.8%) "benign" core biopsies, which contained tissue calcifications > 100 microm and at that time were considered representative of the mammographic lesion, have had a malignant outcome on clinical and mammographic follow-up ranging from 2.4 to 7.5 years. Of the 51 (31%) core biopsies where calcifications were not seen on specimen radiographs, histological calcifications were not found in 34 (67%) core biopsies, whereas in 17 (33%) core biopsies, calcifications measuring < 100 microm were found. All of these core biopsies were considered non-diagnostic and therefore not representative of the lesion targeted. Five (9.8%) of these cases had a malignant outcome with either immediate rebiopsy or excision. Accurate diagnosis of all mammographic lesions requires radiological-pathological correlation. This study shows that the presence of calcifications on the specimen radiograph and the demonstration of tissue calcifications > 100 microm are an essential and highly reliable part of core biopsy assessment for mammographically "suspicious" calcifications. Nevertheless, lesions with "highly suspicious" calcifications on mammography should be considered for excision even if the core biopsy diagnosis is benign and calcifications > 100 microm are present.