© 2000 Blackwell Science Inc., 1075-122X/00/$15.00/0The Breast Journal, Volume 6, Number 5, 2000 294–298

Imaging Guided Biopsy:

An Alternative to Surgical Biopsy

D. David Dershaw, MD

Department of Radiology, Memorial Sloan-Kettering Cancer Center, New York, New York

T

he increasing use of percutaneous large core needlebiopsy of the breast using stereotactic or sonographic

guidance as an alternative to surgical biopsy is supportedby the decreased cost, increased ease, and diagnostic accu-racy of this procedure. Use of these procedures is optimizedwith appropriate patient selection and appropriate ap-plication of biopsy results to patient management.

Stereotactic breast biopsies are done with X-ray im-aging. Because X-rays are used, any type of breast lesionthat can be imaged mammographically can potentiallybe seen with stereotactic equipment. Lesion localizationis done by triangulation. The distance the lesion moveswhen the X-ray beam is tilted is a function of its depth inthe breast, making it possible for a computer to calcu-late this position when angled images are obtained.

Depending on the type of stereotactic equipment used,some lesions will be more difficult to target. When dedi-cated, prone tables (rather than add-on units) are used,lesions that are near the chest wall or in the axilla may bedifficult to include in the field of view of the unit, andthus they may not be able to be targeted. If digital imag-ing is used, some faint microcalcifications may also bedifficult to image. Because of the small field of view, ar-eas of architectural distortion and asymmetric densitycan also be difficult to image; these types of lesions maybe more reliably biopsied using surgery with preopera-tive localization rather than imaging guided biopsy.

Sonographically guided biopsies are performed withthe same equipment needed to do diagnostic sonogra-phy. Therefore acquisition of special imaging equipmentis not necessary. Because calcifications and some massesare not reliably imaged sonographically, the patientpopulation available for sonographically guided proce-dures is smaller than that for biopsies done under stereo-

j

Abstract:

Imaging guided percutaneous core needle bi-opsy of the breast is an accurate and cost-effective methodfor the diagnosis of indeterminate breast lesions. It is alsouseful for the preoperative confirmation of suspected carci-noma. Depending on the location of the lesion in the breast,its imaging characteristics, and the configuration of thebreast, biopsy using sonographic or stereotactic guidancemay be preferable. However, for many lesions either tech-nique can be used. A variety of biopsy probes are available.For many lesions the ability to make a diagnosis does not ap-pear to depend on the probe size or configuration. How-ever, when calcifications undergo biopsy, the retrieval oflarger volumes of tissue may improve the accuracy of the di-agnosis of atypical ductal hyperplasia (ADH), ductal carci-noma in situ (DCIS), and other rare lesions. Major complicationsare rare with these techniques. With benign histologies,most patients can avoid surgery and return to 6-month follow-up or routine screening mammography. When benign histol-ogies are not concordant with the imaging pattern, whencertain high-risk lesions are found at core biopsy, and whenthe pathologist is unable to make a definitive diagnosisbased on the small volume of tissue removed, surgical biopsyis necessary. However, the vast majority of patients with be-nign diagnoses can avoid surgery with these biopsy tech-niques. It should be remembered that these techniques arenot appropriate for the treatment of breast carcinoma atthe present time.

j

Key Words:

breast biopsy, breast neoplasms, breast sur-gery, diagnosis, ultrasound

Address correspondence and reprint requests to: D. David Dershaw,MD, Room A259, Department of Radiology, Memorial Hospital, 1275 YorkAve., New York, NY 10021, U.S.A., or e-mail: dershawd@mskcc.org

Imaging Guided Biopsy

•

295

tactic guidance. However, the entire breast is readily im-aged. Therefore lesions near the chest wall and in theaxilla are easily targeted.

Multiple studies have reported considerable reduc-tion in the cost of diagnosis and the cost of screeningmammography programs by the use of imaging guidedpercutaneous large core needle biopsy rather than surgi-cal biopsy (1–5). These have demonstrated up to a 50%reduction in the cost of diagnosis versus surgical biopsy.In addition, the cost of mammographic screening pro-grams for women beginning at age 40 can be reduced byup to 23%. In some settings it is possible for the patientto undergo biopsy on the day of her imaging examina-tion. Even when the biopsy is prescheduled, the need forvisiting a surgeon and returning after surgery for sutureremoval and other follow-up is eliminated. In addition,because only a small volume of tissue is removed, post-biopsy cosmetic deformity does not occur, as it can withsurgery, and there is no parenchymal scarring seen onthe mammogram that can be mistaken for cancer, re-sulting in additional biopsies in the future.

Complication rates are low with these procedures. Ina multi-institutional series, major complications occurredin 0.2% of women, equally divided between hemor-rhage and infection (6). Minor complications have beenreported in up to 50% of women undergoing these pro-cedures (7). These include ecchymoses, breast pain, andemotional stress. When biopsies are done under sono-graphic guidance, it is possible to violate the chest walland cause a pneumothorax if the needle is incorrectlypositioned before the biopsy gun is fired.

PATIENT SELECTION

Cost-effectiveness and patient satisfaction are opti-mized when appropriate patients are selected to undergoimaging guided breast biopsy. Selection is based on theability to successfully target the lesion, breast size andconfiguration, coexisting medical conditions, and thelikelihood that the lesion is malignant.

As noted above, under imaging guidance areas of ar-chitectural distortion and asymmetry may be difficult tar-gets due to the inability to appreciate these lesions when asmall area of the breast is imaged, as occurs during thesebiopsy procedures. Faint calcifications can be particularlydifficult to appreciate with digital imaging systems avail-able on some stereotactic units. Dedicated, prone stereo-tactic tables often have difficulty including the areas ofthe breast near the chest wall and the axilla in the field ofview. Core biopsy probes require adequate breast thick-ness to accommodate the needle and, if the needle moves

within the breast during tissue acquisition, to accommo-date the needle “throw.” Therefore thin breasts and thinareas of the breast may not be amenable to core biopsy,particularly with stereotactic guidance because the breastis compressed during these procedures.

Under sonographic guidance, some masses and essen-tially all calcifications are not reliably seen and cannotbe targeted. Because these biopsies are done with the pa-tient supine and the breast flattened against the chestwall, breast thickness is not a problem.

The cost-effectiveness of these techniques is based ontheir ability to eliminate the need for surgery in womenwith indeterminate mammographic lesions. Thereforethe use of stereotactic biopsy for women with BI-RADScategory 4 lesions (indeterminate requiring biopsy butprobably benign) accounts for much of the cost savingsfrom imaging guided biopsy. In these lesions, whichhave a less than 50% likelihood of malignancy, the corebiopsy diagnosis of a benign lesion that is consistentwith the imaging pattern of the lesion can eliminate theneed for further intervention in these women (8). Inwomen who have probably benign lesions with a lessthan 2% chance of malignancy and for whom short-term mammographic follow-up is recommended (BI-RADS category 3 lesions) but the patient requests bi-opsy, the use of percutaneous biopsy can also replacesurgical biopsy for diagnosis.

Lesions that have a 75% or greater likelihood of ma-lignancy (BI-RADS category 5 lesions) will almost al-ways require surgery. If a two-stage surgical procedureis planned (an initial diagnostic surgery followed by asecond, therapeutic procedure), the initial surgical bi-opsy can be replaced with a needle biopsy, decreasingthe cost of care. If the patient’s surgery is planned as aone-stage procedure (surgical biopsy immediately fol-lowed by definitive surgical treatment), the use of imag-ing guided biopsy will increase the cost of patient care.However, even in this setting many patients and sur-geons find that the preoperative biopsy diagnosis of car-cinoma by core biopsy facilitates discussions about treat-ment options with the patient and her family. It alsomakes it possible to plan preoperative surgery of theaxilla and increases the likelihood that margins will benegative at the time of the initial excision if breast con-servation is planned (9,10).

In the patient in whom multiple lesions are present inthe breast, the determination of the need for mastec-tomy may depend on the histology present at multiplesites. The presence of multicentric or multifocal carci-noma may require mastectomy, whereas a single site of

296

•

dershaw

malignancy can be treated with conservation. Imagingguided biopsy can be useful in determining the histologyat multiple sites within the breast so that a decisionabout the need for mastectomy can be made preopera-tively (11,12).

Certain medical conditions may be considered con-traindications to the performance of these breast biopsyprocedures. Patients need to be able to give consent forthe procedure and cooperate during it. If this is not pos-sible, it may not be possible to perform these biopsies.Neuromuscular disorders may make it impossible for awoman to stay immobile during the biopsy. Since this isnecessary to maintain the accuracy of the calculation ofthe position of the targeted lesion within the breast, es-pecially during stereotactic procedures, it may not bepossible to do these biopsies on these patients. When us-ing dedicated, prone stereotactic biopsy tables, severearthritis in the neck, back, or shoulders can make pro-longed immobility on these units extremely uncomfort-able and may compromise the ability to perform biop-sies on these women. Coagulation disorders can result inclinically significant hemorrhage. In women with coagu-lopathies, whether inherent or iatrogenic, it may be ad-visable to delay performance of the biopsy until the co-agulopathy can be corrected.

HISTOLOGIC DIAGNOSIS BASED ON CORE BIOPSY SPECIMENS

After the biopsy has been performed, the ability of im-aging guided core biopsy to replace surgical biopsy in theindividual woman will finally depend on the histologicresults in the individual case. For most histopathologies,14-gauge or larger biopsy specimens are adequate tomake an accurate diagnosis. This is almost always thecase for uncalcified masses, which are usually histologi-cally homogeneous. For these lesions a single core of tis-sue through the lesion is often adequate to make an accu-rate diagnosis. However, in lesions that are characterizedby histologic heterogeneity, sampling of the least aggres-sive part of the lesion may result in underestimation ofthe clinical seriousness of the lesion. There are also somelesions for which the removal of a small amount of tissueis not adequate for a definitive diagnosis to be made. Itshould be understood that the success of these proce-dures will vary with the size of the biopsy probe and theability of the pathologist interpreting the specimen.

Atypical ductal hyperplasia (ADH), ductal carcinomain situ (DCIS), and invasive ductal carcinoma are part ofa continuum of alternation of the ductal epitheliumfrom normal to invasive cancer. Therefore these entities

may coexist in the same lesion. Because these can allpresent as an identical pattern of pleomorphic microcal-cifications on mammography, it is impossible to selectas a target the most aggressive site within the lesion atthe time of core, usually stereotactic, biopsy. As largervolumes of tissue are removed, the likelihood of remov-ing an area that contains the most aggressive part of thelesion increases. Therefore accuracy in the diagnosis ofthese lesions at core biopsy increases with increasingcore size and number. Studies based on 14-gauge gun-needle devices indicate that when ADH is obtained atcore biopsy, approximately 40% of patients (reportedrange 20–56%) will have DCIS discovered when the le-sion is surgically reexcised (13–17). Because of this inci-dence of coexistent malignancy with ADH, all of theselesions diagnosed at 14-gauge gun-needle core biopsyshould be sent to surgical reexcision to assess for possi-ble coexistent carcinoma. When larger tissue volumesare excised percutaneously by core biopsy, the likeli-hood of coexistent carcinoma appears to decrease. Eleven-gauge vacuum suction devices decrease the incidence ofunderestimation of DCIS as ADH by about half in mostseries, but at least one author (16) has suggested noappreciable change in underestimation. Therefore, evenwhen larger biopsy probes are used, a histologic diagno-sis of ADH should result in surgical reexcision of thelesion.

Unfortunately the same issues arise with the core bi-opsy diagnosis of DCIS. Underestimation using 14-gauge gun-needle devices results in the discovery of in-vasive carcinoma in approximately 20% of women inwhom a diagnosis of DCIS is made using this type of bi-opsy probe (17–22). When larger volumes of tissue areremoved using vacuum suction, directional biopsy probes,this percentage is decreased by about one-half. Sinceall of these women will go to surgery for reexcision ofthe site of their cancer, this underestimation does notchange the recommendation for surgery. However, insome women this may result in the failure to appreciatethe need for surgical intervention in the axilla at the timeof the original surgery. Therefore a second surgical pro-cedure may need to be done in some patients to surgi-cally assess the axilla after the diagnosis of invasive car-cinoma has been made.

In some less common lesions, reexcision may be nec-essary because of the inability of the pathologist to makea definitive diagnosis based on the small volume of tis-sue excised at core biopsy. It is sometimes not possibleto differentiate phyllodes tumors from fibroadenomasbased on core biopsy; when this problem arises, surgical

Imaging Guided Biopsy

•

297

excision of the lesion is necessary (13). Radial scar is re-ported to be associated with carcinomas (DCIS and tu-bular carcinomas) in a variable percentage of patients. Ifthe diagnosis of this lesion is made on core biopsy, reex-cision is commonly performed (23). Issues regarding theaccuracy of the diagnosis of papillary (24) and muci-nous lesions have also been raised, and surgical excisionof these lesions may also be desirable.

Limited data are available on the need for the reexci-sion of atypical lobular hyperplasia (ALH) and lobularcarcinoma in situ (LCIS) (25,26). With LCIS, it is impor-tant to remember that it is usually mammographicallyoccult, and therefore usually has no mammographicfindings. Because of this, the diagnosis of LCIS usuallydoes not explain the lesion that has been targeted for bi-opsy. Without another reason for reexcision, the discov-ery of LCIS probably does not require that the biopsysite be reexcised, but only a small number of cases tosupport this conclusion have been reported. A similarlack of data surrounds the management of women witha core biopsy diagnosis of ALH.

It is important in all cases in which a benign diagnosisis obtained at core biopsy that the results be correlatedwith the imaging pattern of the targeted lesion. Histo-logic-radiologic correlation is necessary as the final con-firmation that the lesion in question was successfully bi-opsied. In cases in which the histology is not concordantwith the pattern of the suspicious lesion, it should be as-sumed that the lesion was missed at the time of biopsy,and repeat biopsy (either imaging guided or surgical)should be performed. The incidence of repeat biopsythat occurs in any facility’s series will be a function ofthe skill of the physicians performing the biopsy, theskill of the pathologists interpreting the slides, and theincidence of ADH and other lesions requiring repeat bi-opsy in the patient population of the facility. Withinpublished series, rebiopsy rates have ranged from 9 to18% of women undergoing core biopsy procedures(13,15,27). Physicians should be aware that the majorreason for patient dissatisfaction with these biopsy pro-cedures has been the need for repeat biopsy (7). There-fore women should be made aware of this possibility be-fore they consent to these procedures.

PATIENT MANAGEMENT AFTER CORE BIOPSY OF BENIGN ENTITIES

The follow-up interval after core biopsy of benign en-tities has varied at different facilities. While some siteswill recommend 6-month mammographic follow-up ofall women, others return most patients to routine, an-

nual mammographic screening (1,23). It is difficult to as-certain from published series the time at which cancersmissed at false-negative core biopsies were discovered atfollow-up. The rate of missed carcinoma at core biopsyhas been reported at 0–8% (17,28). Some of these havebeen discovered at short-term mammographic follow-up, and others have not shown an interval change untillater follow-up. However, whatever recommendationsare given to patients, they frequently fail to return forfollow-up at the recommended interval (29).

CORE: DIAGNOSIS NOT THERAPY

It is important to remember that core biopsy tech-niques are only approved by the U.S. Food and DrugAdministration for the diagnosis of breast disease. Theiruse as therapeutic techniques is not approved, and datado not support the use of the presently available instru-mentation for the treatment of cancer. It has been dem-onstrated that the extent of malignancy is often greaterthan the volume of the mammographic findings (30). Inone series using 11-gauge vacuum-suction biopsy probes,only 20% (10 of 51) of infiltrating carcinomas werecompletely excised at core biopsy with no residual tu-mor present at surgical reexcision (31). In this series,even when the entire mammographic lesion was re-moved, only 35% of carcinomas had no residual diseaseat surgical reexcision. In another study using 14-gaugevacuum-suction, directional biopsy probes, 13–48% ofmammographic lesions could be completely excised,and 58–93% of targeted lesions measuring 5 mm or lesswere completely removed, also suggesting the inabilityof this technology to therapeutically remove carcinomas(32). Even when the largest of all percutaneous biopsyprobes has been used, the ABBI® device, tumor hasbeen present at the margins of the specimen in 64–100%of cases in reported series.

Future developments may make it possible to treatsome breast cancers by percutaneous imaging guidedtechniques. However, at the present time this technol-ogy is well established for the diagnosis of many breastlesions requiring biopsy with less cost, deformity, andinconvenience than surgical biopsy.

REFERENCES

1. Liberman L, Fahs MC, Dershaw DD,

et al.

Impact ofstereotaxic core biopsy on cost of diagnosis.

Radiology

1995;195:633–37.

2. Liberman L, Feng TL, Dershaw DD, Morris EA,Abramson AF. Ultrasound-guided core breast biopsy: utilityand cost-effectiveness.

Radiology

1998;208:717–23.

298

•

dershaw

3. Lindfors KK, Rosenquist CJ. Needle core biopsyguided with mammography: a study of cost effectiveness.

Radi-ology

1994;190:217–22.4. Lee CH, Egglin TIK, Philpotts LE, Mainiero MB, To-

cino I. Cost-effectiveness of stereotactic core needle biopsy:analysis by means of mammographic findings.

Radiology

1997;202:849–54.5. Fajardo LL. Cost-effectiveness of stereotaxic breast

core needle biopsy.

Acad Radiol

1996;3:521–23.6. Parker SH, Burbank F, Jackman RJ,

et al.

Percutaneouslarge-core needle breast biopsy: a multi-institutional study.

Ra-diology

1994;193:359–64.7. Dershaw DD, Caravella BA, Liberman L. Limitations

and complications in the utilization of stereotaxic core breastbiopsy.

Breast J

1996;2:13–17.8. Smith DN, Christian R, Meyer JE. Large-core needle

biopsy of nonpalpable breast cancers: the impact on subse-quent surgical excision.

Arch Surg

1997;132:256–59.9. Liberman LL, LaTrenta LR, Dershaw DD,

et al.

Impactof core biopsy on the surgical management of impalpablebreast cancer.

AJR

1997;169:495–99.10. Liberman L, Cody HS III, Hill ADK,

et al.

Sentinellymph node biopsy after percutaneous diagnosis of nonpalpa-ble breast cancer.

Radiology

1999;211:835–44.11. Rosenblatt R, Fineberg SA, Sparano JA, Kaleya RN.

Stereotactic core needle biopsy of multiple sites in the breast:efficacy and effect on patient care.

Radiology

1996;201:67–70.12. Liberman L, Dershaw DD, Rosen PP, Morris EA, Co-

hen MA, Abramson AF. Core needle biopsy of synchronous ip-silateral breast lesions: impact on treatment.

AJR

1996;166:1429–32.

13. Dershaw DD, Morris EA, Liberman L, Abramson AF.Nondiagnostic stereotaxic core breast biopsy: results of rebi-opsy.

Radiology

1996;198:323–25.14. Jackman RJ, Burbank F, Parker SH,

et al.

Atypical duc-tal hyperplasia diagnosed at stereotactic breast biopsy: im-proved reliability with 14-gauge, directional, vacuum-assistedbiopsy [abstract].

Radiology

1997;205:325.15. Philpotts LE, Shaheen NA, Carter D, Lange RC, Lee

CH. Comparison of rebiopsy rates after stereotactic core needlebiopsy of the breast with 11-gauge vacuum suction probe versus14-gauge needle and automatic gun.

AJR

1999;172:683–87.16. Brem RF, Behrndt VS, Sanow L, Gatewood OMB.

Atypical ductal hyperplasia: histologic underestimation of car-cinoma in tissue harvested from impalpable breast lesions us-ing 11-gauge stereotactically guided directional vacuum-as-sisted biopsy.

AJR

1999;172:1405–7.17. Liberman L. Percutaneous imaging-guided core breast bi-

opsy: state of the art at the millennium.

AJR

2000;174:1191–99.18. Meyer JE, Smith DN, Lester SC,

et al.

Large-core nee-dle biopsy of nonpalpable breast lesions.

JAMA

1999;281:1638–41.

19. Burbank F. Stereotactic breast biopsy of atypical ductalhyperplasia and ductal carcinoma in situ lesions: improved ac-curacy with a directional, vacuum-assisted biopsy instrument.

Radiology

1997;202:843–47.20. Won B, Reynolds HE, Lazaridis CL, Jackson VP. Ster-

eotactic biopsy of ductal carcinoma in situ of the breast usingan 11-gauge vacuum-assisted device: persistent underestima-tion of disease.

AJR

1999;173:227–29.21. Liberman L, Dershaw DD, Rosen PP,

et al.

Stereotaxiccore biopsy of breast carcinoma: accuracy at predicting inva-sion.

Radiology

1995;194:379–81.22. Jackman RJ, Burbank FH, Parker SH,

et al.

Accuracyof sampling ductal carcinoma in situ by three stereotacticbreast biopsy methods [abstract].

Radiology

1998;209:197–98.

23. Jackman RJ, Nowels KW, Rodriguez-Soto J, MarzoniFA, Finkelstein SI, Shepard MJ. Stereotactic, automated, large-core needle biopsy of nonpalpable breast lesions: false-negativeand histologic underestimation rates after long-term follow-up.

Radiology

1999;210:799–805.24. Liberman L, Bracero N, Vuolo MA,

et al.

Percutaneouslarge-core biopsy of papillary breast lesions.

AJR

1999;172:331–37.

25. Liberman L, Sama M, Susnik B,

et al.

Lobular carci-noma in situ at percutaneous breast biopsy: surgical biopsyfindings.

AJR

1999;173:300–302.26. Lechner MC, Jackman RJ, Parker SH,

et al.

Lobularcarcinoma in situ and atypical lobular hyperplasia at percuta-neous biopsy with surgical correlation: a multi-institutionalstudy [abstract].

Radiology

1999;213:106.27. Meyer JE, Smith DN, Lester SC,

et al.

Large-core nee-dle biopsy: nonmalignant breast abnormalities evaluated withsurgical excision or repeat core biopsy.

Radiology

1998;206:717–20.

28. Parker SH, Lovin JD, Jobe WE, Burke BJ, Hopper KD,Yakes WF. Nonpalpable breast lesions: stereotactic automatedlarge-core biopsies.

Radiology

1991;180:403–7.29. Goodman KA, Birdwell RL, Ikeda DM. Compliance

with recommended follow-up after percutaneous breast corebiopsy.

AJR

1998;170:89–92.30. Dershaw DD, Drossman S, Liberman L, Abramson AF.

Assessment of response to therapy of primary breast cancer bymammography and physical examination.

Cancer

1995;75:2093–98.

31. Liberman L, Zakowski MF, Avery S,

et al.

Completepercutaneous excision of infiltrating carcinoma at stereotacticbreast biopsy: How can tumor size be assessed?

AJR

1999;173:1315–22.

32. Liberman L, Dershaw DD, Rosen PP, Morris EA,Abramson AF, Borgen PI. Percutaneous removal of malignantmammographic lesions at stereotactic vacuum-assisted biopsy.

Radiology

1998;206:711–15.