The American Journal of Surgery © Volume 168 December 1994

B ACKGROUND:Stereotactic localization and breast biopsy by fine needle and 14-gauge core needle is a new technique for diagnosing nonpalpable breast lesions. The procedure em-ploys a device that affords extremely accurate localization and sampling of nonpalpable breast abnormalities.

METHODS: We are a 5-man surgical group re- porting on the experience of our initial 300 consecutive stereotactic core-needle biopsies (SCNB).

RESULTS:The procedures, conducted over a 13-month period, revealed 37 cancers, for a malignancy rate of 12%. Seven percent were infiltrating- ductal carcinoma and 4% ductal carcinoma in situ. There were 2 cases of adenocarcinoma, and I case of mucinous carcinoma. Benign microscopic diagnoses included 193 categorized as "fibrocystic change," 34 "fibroadenomas," 19 "benign breast tissue" and 5 lesions that were suspected of being malignant but were proven to be benign. There were 12 "others."

CONCLUSION:We conclude that SCNB is an essentially painless, short outpatient procedure with a reduce cost compared to open biopsy. It can be easily mastered by surgeons. Results are comparable to controlled series in the literature, and rates of malignant diagnosis are to our group's experience in previous years.

The emphasis on early detection and prevention of breast cancer has dramatically increased the number of mammograms performed. As a result, many mammographically worrisome nonpalpable breast lesions are being detected.

Stereotactic core-needle biopsies (SCNBs) were per- formed when a mammographer interpreted a film as showing a suspicious lesion and recommended biopsy. The surgeon performing the procedure always reviewed the film and confirmed that the lesion was appropriate for SCNB. The surgeon classified each case prior to biopsy. Class I included normal, benign mammograms, none of which were biopsied. ClassII comprised lesions that
were felt to be most likely cysts, lymph nodes, fibroadenomas, or other benign growths. They were scheduled for biopsy because they changed mammographically or had worrisome characteristics, or because the patient wished it Class III lesions were those that were considered about equally likely to be malignant or nonmalignant, including both solid lesions and suspicious calcifications. Class IV lesions had a high suspicion of malignancy. This class assignment by the surgeon was independent of the mammography report, and was not standardized.


The device used for the procedure was the Fischer Imaging Mammotest with Auto-Guide andthe Mammoscan System (Fischer, Inc., Denver, Colorado), which allows for digital imaging in just a few seconds following roentgenographic exposure. With the patient prone on the table, a O digital scout image was obtained, followed by 2 views, each 15 off center line. Using the screen images, biopsy coordinates and depths for six to eight cores were
calculated by the computer. A 14-gauge core needle was into the breast, and stereotactic images confirmed the pre- and postfire I on for the first biopsy site. The cores of tissue were removed by the Biopty-Cut Biopsy Needle (Bard Inc., Murray Hill, New Jersey). Specimen roentgenograms were performed for calcifications in some cases and, on occasion, frozen section. Cores were then submitted for pathology m formalin for routine tissue processing.


The 300 lesions biopsied were from 288 patients, some of whom had bilateral lesions and others distinctly separate (distant) lesions in one breast. No patient received a diagnosis of bilateral malignancies, and in no case were two separate lesions within the same breast both malignant.


The patients ranged in age from 30 to 89 years, with a median of 55. One hundred fourteen patients (40%) had family histories of breast malignancy. Of these, 18 (16%) received malignant diagnoses from SCNB of their lesion. The median number of cores taken from a patient was 6 (range I to 15). In 24 cases, the digital scout image failed to show the lesion that was deemed suspicious on mammography. During the early months of the study, patients were immediately reimaged with standard film techniques when this occurred. The correlation between the Mammotest image and the standard film image was complete, so the repeat mammography was eventually dropped. By the time we had done
approximately 100 cases, the procedure time from initial image-taking to completion of the stereotactic core-needle biopsy tissue acquisition seldom exceeded 30 minutes. For lesions that were centered on the initial image, acquisition times were 15 to 20 minutes. All images were stored on a computer hard disk until completion of the day's cases and then archived with a laser- imager device on conventional roentgenographic film.


Patients were contacted a few days after their biopsies to discuss the pathology
results and evaluate the biopsy sites. All were scheduled for examination and follow-up mammography between 4 and 6 months.

In the final pathology reports,.263 lesions (88%) were diagnosed as benign and 37 (12%) as malignant. Twenty- two of the malignancies were infiltrating ductal carcinomas, 2 adenocarcinonias, and 1 a mucinous carcinoma. Twelve were ductal carcinoma in situ, 2 of which later proved on excisional biopsy to have invaded surrounding tissue. Three of the infiltrating ductal cancers were 4 mm or less in size. There were no lobular carcinomas insitu. Six (16%) of the malignancies were in patients aged 50 or younger.


The pathologists correctly diagnosed 258 of the 263 (98%) benign lesions and 31 of the37 (84%) malignant lesions based on the SCNB specimens.They adjudged 11 "samples suspicious for malignancy," including atypia and radial scar. All the "suspicious" lesions were investigated by needle localization and excisional biopsy. In the final pathology reports, 4 were infiltrating ductal carcinoma, 2 ductal carcinoma in situ, and 5 benign &"fibrocystic change."

Other pathology reports from the core biopsies are shown in Figure 1.In three cases, the tissue obtained in SCNB was inadequate to make a diagnosis. AD threewere followed- lowed up with needle localizations and excisional biopsies, and none weremalignant Although ultrasound was employed whenever mannnographic lesions looked like cysts, three lesions that we attempted to biopsy turned out to be cysts that ultrasound had misdiagnosed. We drained them with the core needle. They are not counted among the 300 SCNBs reported here, all of which are actual core-tissue biopsies.

Table II lists the results for the SCNBs by the class of lesion assigned by the surgeon. >

Our surgical group treated 99 breast cancers during calendar year 1992 and 96 during1993. Calendar year 1993 essentially corresponded to the first I 11 of the 13 months required for performance of the 300 core biopsies. A trend toward treating cancers in less advanced stages began between 1992 and 1993. The number of stage 0 malignancies doubled,stage I increased by a third and stage II -A by one third. The more advanced stages, II-B through IV, remained about equally numerous Figure 2. The numbers are too small for statistical significance, at this time.


The American Joint Committee on Cancer Staging System (4th edition) was used for all cases. Ten (27%) of our SCNB-diagnosed malignancies were stage 0, 22 (60%) stage I, 2 (5%) stage II-A, 2 (5%) stage II-B, and 1 (3%) stage IV.

The total cost of the SCNB procedure averaged $1,270 per patient. The average total cost for an outpatient freehand needle localization followed by open breast biopsy unde local anesthesia was $2,450. The cost for outpatient open breast biopsy under general anesthesia was significantly higher.
Of the first 139 patients who had a benign SCNB, 127 have had mammograms 4 to 6 months postbiopsy. None of the biopsied lesions have demonstrated growth or increasing suspicion requiring additional investigation. We are currently contacting the remaining 12 patients and asking them to return for mammograms.

SCNB has not resulted in any complications requiring treatment. An extremely small number of patients have m- ported discomfort requiring mild analgesics. Them have been no major hematomas requiring drainage and no reported infections.

Numerous studies have concluded that fine-needle aspiration biopsies and stereotacticcore biopsies are as accurate as needle localization and open surgical biopsy for the diagnosis of clinically occult breast malignancy seen on mammography. Stereotactic core-needle breast biopsy entails much less patient discomfort and disfigurement than surgical biopsy. With experience, it can be performed in less time and at approximately half the cost. It can now be accomplished in an outpatient setting with essentially nocomplications.


Experts in cytologic sampling suggest that the aspiration and smearing of fine-needleaspirates are an "art," and that the interpretation requires a very experiencedcytologist, as no histologic information is provided with the specimen. However, in large series of stereotactically guided fine-needle aspirations of nonpalpable breast -lesions, the sensitivities are consistently above 90%, specificities even higher, and positive predictive values surpass 90%, as well.

We are aware of only one case in our institution where the initial SCNB missed amalignant lesion. The misdiagnosis was later recognized and repeat SCNB was positive. This case is not included in this report, as it involved physicians other than our surgical group. We conclude that any physician performing this procedure needs. to be very familiar with the lesion being sampled, on the diagnostic mammogram as well as the digital images. One must be willing to repeatedly reconfirm the lesion when any degree of uncertainty exists concerning the correlation.


It is appropriate to count "suspicious" lesions that turn out to be benign as false-positives, since these patients are subjected to open biopsy along with those who have "malignant" findings. When this is done, SCNB is seen to have correctly identified 258 of 263 benign lesions in this series. The 5 false-positives yield a specificity of 98% for SCNB. If "suspicious" lesions that turn out to be malignant are counted as true-positives, all currently known malignant lesions were diagnosed with SCNB, for a 100% sensitivity rate. Obviously, this rate will need to be confirmed with further follow-up, as some additional lesions may prove to be malignant in the future.


Our observations coincide perfectly with the experience of Mitnick et al,who
found that definitive benign and malignant pathology reports are reliable. Suspicious, or atypical, biopsy results are almost always malignant and require open biopsy. Inadequate tissue may either be repeat biopsied with the SCNB technique or excised.


We recognize the early results of this series do not permit the long-term evaluation ofthe lesions diagnosed as "benign" that may later prove to be malignant. The responsibility of the physician performing these procedures includes careful monitoring of post SCNB mammographic follow-up.


Although the malignancy rate of 12% of lesions biopsied is lower than we would like, the "best" percent of previous years' experience by this group in traditional
needle localization and open breast biopsy was a 10% malignancy rate. It is hoped that positive, continuous feed-back to the mammographers will result in a significantly increased rate.

1. Parker SH, Lovin JD. Jobe WE, et al. stereotactic breast biopsy with a biopsy gun. Radiology. 1990; 176:741-747.

2. Parker SH, Lovin JD. Jobe WF- et al. Nonpalpable breast lesions: c automated large-com biopsies. Radiology. 1991;180: 403-407
3. Elvecrog EL, Lechner MC, Nelson MT. Nonpalpable breast lesions: correlations of stereotactic large-core needle biopsy and surgical biopsy results. Radiology. 1993;198:453-455.

4. Mitnick JS. Vazquez W, Ross DF, et al. S c I on for fine-needle aspiration breastbiopsy. Initial experience with 300 patients. Arch Surg. 199 1; 126:1137-1140.

5. Dow K Gent IU. Schmidt P, et al. Nonpalpable breast tumors: diagnosis with stereotactic localization and fine-needle aspiration Radiology. 1989; 1 70:427-433.

6. Dow K Jokich PM, Kluskens LF, et al. A prospective study of double diagnosis of nonpalpabic lesions of the breast Surg Gynecol Obster. 1991;172:121-124.

7. Gent HJ, Sprenger E, Dowlatshahi K. Stereotwtic needle localization and cytological diagnosis of occult breast lesions. Ann Surg. 1986;204:580-584.

8. Oliver DJ, Frayne JR, Steffett G. S c fine needle biopsy of the brML Aust N Z J Surg. 1992.62:463-467.

9. Schmidt R, Monow M. Bibbo M. Cox S. Benefits of c aspiration cytology. Admin Radiol. 990;Oct:35-42.

10. Kopans DB. Fine-needle aspiration of clinically occult breast lesions. Radiology.1989;170:313-314.

Edward W. Nelson, MD(Salt Lake City, Utah): Our experience with stereotactic core-needle biopsies for nonpalpable mammographic lesions and our initial impression are very similar to those presented by Dr. Janes and his group. This is a technique of some benefit that thus far has belonged exclusively tothe radiologist. Although the surgical literature is very scant regarding this technique, I am very happy to see that this paper describes a series of patients who were followed, evaluated, and treated by surgeons. Series like these will help define the proper indications for this procedure.


It is a good technique, but what are the indications? authors present 300 cases and have broken down their mammograms into classes I, II, III, and IV, depending upon the suspicion of malignancy. The total number of mammograms done during this period is not presented, so we do not know what the percentage of their total experience this represents. Also, specific criteria that distinguished one class of mammogram from another is of critical importance and perhaps could have been presented in more detail.

My questions address the indications for stereotactic guided core biopsies for these various groups. Of the 59 patients with class II, likely benign mammography, only 1 was found to be malignant on Mammotest biopsy. I would like to ask the authors if that group of patients might not be safely followed without any type of biopsy. This has been suggested by other large series in the literature, such as Sickle's, who followed over 3.000 such mammograms. At the other end of the spectrum, the class IV mammograms, which were read as probably malignant, had a 41 % positivity rate. Is Mammotest biopsy really needed for these lesions? Could they not proceed directly with wire localization or Mammotest locations and have an open biopsy/lumpectomy as part of their initial procedure for breast cancer? What's left, of course, are the class III lesions, those that were suspicious but not obviously malignant or benign. This is the most interesting group, and, in our series, we chose to limit Mammotest biopsy to this group. Under such guidelines, our percentage of positivity turned out to be 8.2%, as compared to the authors' 9% for the same mammographic class of lesions.

This brings me to my first of three questions. Based on your preliminary results, would you consider using Mammotest biopsy only for the evaluation of class III, while following those in class II, and going directly to open biopsy for those in class IV? My second question regards follow-up. The authors state that 150 patients who have had this procedure are eligible for follow-up at 6 months. Based on our experience, I am astonished that they really had 100% follow-up with mammography at 6 months. Is this really the case? If not, what is their total percentage of follow-up with mammography at 6 months? Finally, this procedure presumably will diagnose breast cancer at an earlier stage. Can the authors give us any information as to whether breast conservation therapy has been more common since this technique has been used?

I very much enjoyed reading this paper, especially comparing it to our own series. The similarities are striking, especially for patients with suspicious but not obviously malignant or benign mammography. I would emphasize that stereotactic core-needle biopsy is a very useful procedure, and I applaud the fact that it has now been studied and presented by a group of surgeons.


Robert H Janes, MD Let me try to take Dr. Nelson's questions m order First, how many mammograms are done in our area? There are about 10 mammographic sites, all but I of which are approved by the American College of Radiology. There are a lot of different mammographers reading, and I really don't have a handle on the numbers. The answer is, I really don't know what the denominator is and I'm not sure whether I know how to find that out. We looked in our own center, and between the hospital and the clinic that we use, between 40 and 50 mammograms are done per day. However, there are many more done in that area.

Second, can we follow class II? Well, we had to start somewhere. What we were facedwith was biopsying lesions that the radiologists recommended we biopsy. In all of ourcases, the radiologists had recommended a biopsy. We felt uncomfortable because we felt we were doing too many, but we had to start gathering data somewhere if we were ever going to change.

Third, talking about class IV and doing open biopsies, I guess that is a matter of
philosophy. We virtually don't do any treatment of breast cancer without a pretreatment diagnosis. We needle all palpable lesions, either fine-needle or core, in the office. Stage Is, a lot of stage IIs and IIIs, also certainly most stage IVs, are presented or seen by our other specialists, radiotherapists and medical oncologists, where indicated before surgery is done.


Fourth, do we have 100% follow-up on the first 150 cases? No, there are 4 or 5 yet to be contacted. We are in a large clinic with a very sophisticated management informationsystem, and each patient, once they had an SCNB, was given an appointment for either 4 or 6 months and a mammogram was scheduled for that day. If they don't show. or if they cancel that appointment, their name is printed out automatically, a letter is put on our desk, and it is sent out. If they don't reschedule within a few weeks, calls are made. We believe you must set up some kind of system to follow these patients. The question mark in
our specificity and sensitivity means that we have to follow these patients carefully.


The fifth question asks if there is an increased number of conservation operations. I showed you that in a slide, but I didn't show you comparison figures. We doubled our stage 0s and we increased, over a third, our stage Is this last year. Unless they refuse, these patients are seen by a radiotherapist and/or a plastic surgeon after diagnosis and before treatment. There are a lot of logistic problems that have to do with the rural area.

The last question, is estrogen/progesterone available on a core biopsy? Yes. Regarding the patient with the solid lesion I showed on the slide, we did perform ER/PR studies on her. She waited about 3 weeks for treatment while going to Mayo Clinic for a second opinion. She had all those studies done, so you certainly can do them. However, most of the time we will wait until we do definitive treatment. This patient had very favorable prognostic studies.
Webmaster@earthlinks.com