Comparison of stereotactic fine needle aspiration biopsy and core needle biopsy in breast lesions

Sammanfattning: Purpose: The purpose of this study was to evaluate the diagnostic value of two biopsy methods by comparing stereotactic fine needle aspiration biopsy (S-FNAB) with stereotactic core needle biopsy (S-CNB). The biopsies were performed by the same radiologist, in the same breast lesion on the same occasion, in order to establish which method that best can be recommended in different kind of mammographically detected lesions. All lesions that were surgically extirpated after the stereotactic biopsies and had final post-operativ pathological anatomical diagnoses (PAD) were included in this study. Introduction: In Sweden fine needle aspiration biopsy (FNAB) has been successfully used in breast lesions since the 1960th due to both excellent biopsy technique and cytologists. Cytology is not standard procedure in many countries, due to lack of experienced cytologic diagnosticians and incorrect biopsy technique. Thus surgical biopsies have been performed instead of needle biopsies in many places. During the 1970th core needle biopsy (CNB) was developed and the method has later been refined. When we started our study 1993, CNB was not used in breast lesions in Sweden. Biopsy technique: Spinal needles with diameter 0.7 and 0.9 mm were used for S-FNAB and an average of three needle biopsies per lesion were taken. For S-CNB a needle with diameter 2.1 mm and 23 mm throw was used. Three needle biopsies per lesion were taken. Paper I: Between May 1993 and June 1998 pre-operative S-FNAB and S-CNB were performed on a single occasion in 22 breast lesions where post-operative PAD showed invasive lobular cancer (ILC). S-FNAB diagnosed cancer in nine (41 %) and probable cancer in five of the 22 lesions. In three cases S-FNAB showed atypia and in five normal cells without any atypia. S-CNB diagnosed ILC in 20 (91 %) of the 22 patients, a mixture of ILC and invasive ductal cancer (IDC) in one lesion and ductal cancer in situ (DCIS) in the final lesion. In conclusion S-CNB was superior to S-FNAB in lesions where post-operative PAD diagnosed ILC. Paper II: In 72 breast lesions that pre-operatively underwent simultaneous S-FNAB and S-CNB between May 1993 and June 1999 post-operative PAD diagnosed ductal cancer in situ (DCIS). S-FNAB in these 72 lesions diagnosed cancer in 34 cases (47 %) and probable cancer in six lesions, atypia in 12 cases and in 20 tumors the material was benign or unsatisfactory. S-CNB in the same lesions performed on the same occasion diagnosed DCIS in 56 cases (78 %). Another three biopsies showed probable cancer, seven showed atypia and in six lesions only benign material was found. In four of the 72 lesions (6 %) S-FNAB was superior to S-CNB and diagnosed cancer. S-CNB in these four lesions diagnosed two probable cancer, one atypia and one with benign material. In another four cases both methods showed only benign material and in four lesions both methods found atypia. In all of these 12 cases the radiologist had recommeded surgical extirpation due to the suspicious mammographic appearance. Paper III: From May 1993 to December 2000 522 patients underwent surgical extirpation of a breast lesion after simultaneous pre-operative S-FNAB and S-CNB. In 448 of these cases post-operative histopatology diagnosed malignancy and in 74 a benign lesion. S-FNAB pre-operatively diagnosed 254 of the 448 cancers (57 %) and in 48 cases diagnosed probable cancer. S-CNB diagnosed 388 of the cancer cases (87 %) and in 18 probable cancer. S-FNAB was false negative in 96 patients (21 %), while S-CNB was false negative in 22 cases (5 %). In 16 of the 74 benign breast lesions (21 %) PAD diagnosed radial scar, which is considered pre-malignant. Paper IV: Between September 1994 and December 2000 three S-CNBs were taken from every lesion irrespectively of its mammographical appearance. The lesions were divided into three groups depending on their mammographical appearance; microcalcifications only (group I), a mass and microcalcifications (group II) and a mass, a star or distorsion without microcalcifications (group III). Every biopsy was analysed separately. 523 of these breast lesions were extirpated surgically with a post-operative PAD. 454 lesions were malignant (87 %) and 69 were benign. Three S-CNBs diagnosed malignancy in 84 % of all cases in group I. In group II 97 % got a correct pre-operative malignant diagnosis with three S-CNBs, while the correct diagnosis was made in 93 % in group III. These results indicate that three S-CNBs are enough in group II and group III but not sufficient in lesions with microcalcifications only (group I). In spite of the differences concerning diagnostic accuracy the post-operative PAD ratio malignant to benign in the three groups were essentially the same i.e. approximately 85 % malignant lesions. The reason for this is that the interpretation of the mammograms also has to be taken into account when deciding if a surgical extirpation shall be done or not and not only the cytological and histopathological results of the biopsies. Final conclusions: When only the biopsy methods are compared, the pre-operative diagnostic results are generally better with S-CNB than with S-FNAB, especially in lesions diagnosed as ILC and DCIS. S-FNAB in combination with S-CNB can be valuable, since few lesions are pre-operatively diagnostic only with S-FNAB. In clinical routine, the combination of the mammography and the biopsy methods must be evaluated together. A combination of mammography and three S-CNBs gives the best diagnostic outcome in all types of breast lesions.