Annals of Surgical Oncology

, Volume 12, Issue 2, pp 181–188

Stereotactic Radioguided Surgery by SiteSelect for Subclinical Mammographic Lesions

Authors

    • Department of General and Breast SurgeryInstitut Curie
  • Martine Meunier
    • Department of RadiologyInstitut Curie
  • Carl El Khoury
    • Department of RadiologyInstitut Curie
  • Claude Nos
    • Department of General and Breast SurgeryInstitut Curie
  • Anne Vincent-Salomon
    • Department of PathologyInstitut Curie
  • Brigitte Sigal-Zafrani
    • Department of PathologyInstitut Curie
  • Krishna B. Clough
    • Department of General and Breast SurgeryInstitut Curie
Article

DOI: 10.1245/ASO.2005.01.004

Cite this article as:
Doridot, V., Meunier, M., Khoury, C.E. et al. Ann Surg Oncol (2005) 12: 181. doi:10.1245/ASO.2005.01.004

Abstract

Background

We defined the indications for and evaluated the results of a new technique for radioguided surgery, the SiteSelect system. The procedure allows en-bloc resection of the breast parenchyma under local anesthesia.

Methods

This prospective study was based on 167 patients operated on between December 2000 and October 2003 with 2 phases. The first step was an evaluation of the feasibility of the procedure with the 15-mm cannula, and the second was therapeutic with the 22-mm cannula.

Results

The mean duration of the procedure was 42 minutes. In 96.9% of procedures, the lesion was excised successfully. Only one complication (hematoma) and two failures and were observed. Histological examination revealed benign disease in 65.8% of cases and cancer in 34.2% of cases. In the latter cases, the specimen margins were histologically involved in 86.2% of cases with the 15-mm procedure and in 41% with the 22-mm procedure. During the first evaluation, all patients with a cancer underwent systematic surgical re-excision: residual tumor was present in 18 cases (64.2%). The biopsy was painless for 88 patients, and the cosmetic result was good in all cases.

Conclusions

This study shows that the SiteSelect procedure allows resection of the lesion in 96.9% of cases. Combined with complementary surgical lumpectomy during the same operation, this procedure achieved a success rate of 98.7%. In the case of cancer, the 15-mm cannula is not wide enough to allow free margins. The use of a new 22-mm cannula, currently under evaluation, might solve this problem.

Keywords

Breast cancerMarginsSiteSelectStereotactic surgical biopsy

Mammographic breast cancer screening may demonstrate undetermined or suspicious subclinical lesions that require histological verification. Until the early 1990s, the only available sampling technique was surgical biopsy. The development of percutaneous biopsies has gradually replaced the indications for surgical biopsies because of the introduction of 18-gauge and 14-gauge microbiopsies,1,2 followed by 11-gauge vacuum-assisted core macrobiopsies (Mammotome; Ethicon, Somerville, NJ).3,4 These biopsies allow histological diagnosis of a suspicious mammographic lesion, but when pathologic analysis shows undetermined lesions or breast carcinoma, complementary surgical resection has to be performed. Radioguided breast surgery was described initially in 1996. It allows complete en-bloc resection of a mammographic lesion and provides the pathologist with a nonfragmented biopsy sample. Two systems are currently available: the Advanced Breast Biopsy Instrumentation (ABBI) system (United States Surgical Corporation, Norwalk, CT), the marketing of which was stopped in September 2002, and, more recently, the SiteSelect system (Imagyn Medical Technologies, Irvine, CA). We conducted a prospective study to evaluate the SiteSelect and define its diagnostic and therapeutic indications.

MATERIALS AND METHODS

This study was based on 167 consecutive patients who underwent operation between December 2000 and October 2003. The mean age of the patients was 58.4 ± 11.3 years (range, 35–88 years). Biopsies could not be performed in six cases: microcalcifications were not visible on digital images for three patients, and the compressed breast was too thin in the other three cases. Bilateral biopsies were performed in two cases. Therefore, the same surgeon (V.D.) performed 161 biopsies.

SiteSelect is a stereotactic resection system adapted to a dedicated table (Lorad Medical Systems, Danbury, CT); it allows en-bloc resection of a mammographic lesion and adjacent tissue while preserving subcutaneous tissue. It is composed of a 15- or 22-mm cutting cannula that is equipped with a garrote wire at its tip. The cannula contains a needle into which a hook wire is introduced (Fig. 1). Two systems are connected to either side of the cannula. The first is a screwing system that allows the cannula to cut the breast tissue. The second system is a garrote wire, which sections the deep edge of the biopsy core and frees the specimen at the end of the procedure. After the patient is placed in the prone position, the biopsy system is connected to the stand. Stereotactic localization of the lesion is performed (first image, +15° and −5° views; Fig. 2). Superficial local anesthesia is performed with lidocaine. The SiteSelect needle is then inserted into the target. A second image (+15° and −15° views) is taken to verify correct positioning of the needle in the center of the lesion (Fig. 3). If not correctly placed, the needle is repositioned. The hook wire is then advanced in the needle and released. Local anesthesia is completed, and a skin incision, approximately 20 mm long, is made. The cutting cannula is then advanced, cutting the breast tissue in front of the lesion. A manual rotation movement allows the cutting cannula to section a core on either side of the lesion. Finally, the garrote wire transects the deep edge of the specimen by squeezing two lateral triggers on the device. Two mammograms are taken at the end of the operation, one of the biopsied breast (third image) and one of the specimen (fourth image) (Fig. 4), to confirm resection of the biopsied lesion. The patient is then placed on her back. After verification of hemostasis, by using an electrocautery device if necessary, the breast and skin are sutured in two layers.
https://static-content.springer.com/image/art%3A10.1245%2FASO.2005.01.004/MediaObjects/10434_2005_2404_f1.jpg
Fig 1

A stereotactic resection system adapted toa dedicated table (SiteSelect; Imagyn Medical Technologies, Irvine, CA).

https://static-content.springer.com/image/art%3A10.1245%2FASO.2005.01.004/MediaObjects/10434_2005_2404_f2.jpg
Fig. 2

Stereotactic localization of the lesion(+15° and −15° views, left and right, respectively).

https://static-content.springer.com/image/art%3A10.1245%2FASO.2005.01.004/MediaObjects/10434_2005_2404_f3.jpg
Fig. 3

A second image (+15° and −15° views,left and right, respectively) is taken to verifycorrect positioning of the needle in the center of the lesion.

https://static-content.springer.com/image/art%3A10.1245%2FASO.2005.01.004/MediaObjects/10434_2005_2404_f4.jpg
Fig. 4

A mammogram taken of the specimenat the end of the operation to confirm resectionof the biopsied lesion.

The study was divided into two parts. The first step assessed the feasibility of the procedure with the 15-mm cannula. One hundred thirty-one patients with American College of Radiology (ACR) stage III (n = 60) and ACR stage IV (n = 71) lesions were enrolled in the study. For the ACR stage III lesions, all patients had medical history of familial carcinoma or had a lesion of high breast cancer risk. Our protocol required systematic surgical re-excision under general anesthesia for patients with in situ ductal or invasive carcinoma to verify the quality of the resection. The aim of the second step of the study was therapeutic with a 22-mm procedure. It involved 36 patients with ACR stage IV (n = 28) and ACR stage V (n = 8) lesions. Systematic lumpectomy under local anesthesia and clips in the excision bed were routinely performed at the end of the procedure.

One hundred thirty-four patients (80.2%) had microcalcifications, which were grouped in a single focus in 103 cases (76.9%) and were scattered in 31 cases (23.1%). Thirty-three patients (19.8%) presented with an opacity. The topography of the lesions was distributed as follows: lateral quadrants (55.6%), medial quadrants (18.6%), and other (25.8%).

One third of the patients were reviewed 6 months after the procedure to evaluate the cosmetic result and to perform a control mammography. Both the quality of the scar and the overall cosmetic result were evaluated on a scale of 1 (excellent) to 5 (poor). The patient’s criticisms and level of satisfaction were systematically recorded.

RESULTS

The mean procedure time (from placement of the patient in the prone position until skin suture) was 42 minutes. The incision was performed in the lateral quadrants in 128 cases (77.1%), in the medial quadrants in 21 cases (12.6%), and in another quadrant in 12 cases (10.3%).

The needle had to be repositioned in 45 cases (27.1%). In 156 cases (96.9%), the procedure was uneventful, and the lesion was excised successfully. Surgical hemostasis with electrocautery was necessary before glandular suture in 11 patients (6.8%). Five failures (3.1%) were observed (persistence of the lesion on the mammogram at the end of the procedure or absence of a lesion on the image of the biopsy specimen). For these patients, the initial mammographic lesions were single-focus microcalcifications. Two of these failures were due to a topographical problem: one patient presented with subcutaneous microcalcifications, and the other presented with very deep microcalcifications, close to the pectoralis major. Only one complication occurred in this series: the pectoralis major was lacerated by the cutting cannula, causing an intramuscular hematoma. Hemostasis and surgical drainage under general anesthesia had to be performed immediately after the procedure. Three failures were related to complete blocking of the garrote wire at the end of the procedure without the presence of a lesion. Breast tissue was dense, and the manual rotation of the cannula was difficult. However, we then released the biopsy core adherent to the deep plane by performing manual section with a cold scalpel. A complementary surgical lumpectomy was performed under local anesthesia during the same procedure, thus ensuring resection of the lesion. Another mammogram was taken at the end of the procedure to confirm resection. Thus, only two complete failures were encountered in this series. Combined with completion surgical lumpectomy, 161 biopsies were successfully achieved, for an overall success rate of 98.7%.

We deliberately limited our indications to ACR stage III (n = 60) and ACR stage IV (n = 71) lesions for the evaluation procedure. For the second part, there were 28 ACR stage IV cases and 8 ACR stage V cases. One hundred fifty-five cases (except scattered microcalcifications and the radial scar) could theoretically have been approached by 14-gauge biopsy. However, macrobiopsy by Mammotome is the current standard for focus microcalcifications. The 14-gauge biopsy could have been performed only for mammographic mass except the radial scar (32 cases) because this kind of biopsy is not appropriate for this lesion. Histological results are listed in Table 1. The lesion was related to benign lesions in 106 cases (65.9%), including atypical epithelial hyperplasia in 18 cases (11.2%). Benign lesions were correlated with ACR stage III lesions in 54 cases and with ACR stage IV lesions in 52 cases. In situ carcinoma (ductal or lobular) was found in 30 cases (18.6%) and invasive carcinoma in 25 cases (15.5%). In situ and invasive carcinomas were respectively correlated with 27 and 20 ACR stage IV lesions and with 3 and 5 ACR stage V lesions.
TABLE 1

Histological results (n = 161)

Variable

15-mmcannula

22-mmcannula

n (%)

Benign lesions (n = 106)

   

  Fibrocystic disease

52

8

60 (37.2)

  Fibroadenomas

10

1

11 (6.7)

  Galactophoritis

1

0

1 (.7)

  Cyst

1

1

2 (1.3)

  Benign adenosis

2

0

2 (1.3)

  Epithelial hyperplasia    without atypia

6

0

6 (3.6)

  Radial scar

1

0

1 (.7)

  Papilloadenoma

1

2

3 (1.9)

  Intramammary node

1

1

2 (1.3)

  Atypical ductal hyperplasia

15

3

18 (11.2)

Noninvasive carcinoma (n = 30)

5

3

8 (4.9)

  Lobular carcinoma-in-situ

18

4

22 (13.6)

  Ductal carcinoma-in-situ

   

Invasive carcinoma (n = 25)

11

12

23 (14.2)

  Invasive ductal carcinoma

1

1

2 (1.3)

  Invasive lobular carcinoma

   

Total

125

36

161 (100)

Our protocol required systematic surgical re-excision for patients with in situ ductal or invasive carcinoma to verify the resection quality. Twenty-two wide excisions of the biopsy bed under general anesthesia were therefore performed (this operation could not be performed in two cases because of the patient’s poor general state). Histological results of the comparison of the margins of the initial SiteSelect biopsy and the subsequent wide excision specimen are listed in Table 2. In 86.2% of cases, the SiteSelect biopsy margins were involved with tumor. Residual tumor was present in 64.3% of cases on the re-excision specimen. No residual tumor was observed in patients with uninvolved margins on the initial biopsy.
TABLE 2

Invasive and noninvasive carcinomas (15-mm cannula)

Histological type

Mammographic image

Size (mm)

SiteSelect margins

Reoperation

Presence of residual tumor

IC

Zone of micro (3 mm)

5

Free

Lumpectomy

No

 

Zone of micro (6 mm)

1.6

Free

Lumpectomy

No

 

Zone of micro (6 mm)

9

Involved

Lumpectomy

No

 

Zone of micro (5 mm)

5

Involved

Lumpectomy

No

 

Zone of micro (7 mm)

18

Involved

Lumpectomy

No

 

Zone of micro (18 mm)

13

Involved

Lumpectomy

No

 

Zone of micro (4 mm)

6

Involved

Lumpectomy

No

 

Zone of micro (5 mm)

6

Involved

Lumpectomy

Yes (IC)

 

Zone of micro (10 mm)

15

Involved

Mastectomy

Yes (IC)

 

Zone of micro (5 mm)

6

Involved

Lumpectomy

Yes (IC)

 

Zone of micro (6 mm)

18

Involved

Lumpectomy

Yes (IC)

 

Zone of micro (6 mm)

18

Involved

Lumpectomy

Yes (IC)

 

Zone of micro (6 mm)

12

Involved

Lumpectomy

Yes (IC)

 

Zone of micro (9 mm)

80

Involved

Mastectomy + IBR

Yes (IC)

 

Scattered micro

25

Involved

Mastectomy

Yes (IC)

 

Scattered micro

6

Involved

No second-look surgery

Yes (IC)

 

Scattered micro

50

Involved

Mastectomy

Yes (IC)

 

Scattered micro

50

Involved

Mastectomy + IBR

Yes (IC)

IDC

Opacity (5 mm)

4

Free

Lumpectomy + ALND

No

IDC

Opacity (6 mm)

6

Free

No second-look surgery

Yes (IDC)

IDC

Zone of micro (4 mm)

6

Involved

Lumpectomy + ALND

No

IDC

Zone of micro (5 mm)

6

Involved

Lumpectomy + ALND

Yes (IDC)

IDC

Zone of micro (6 mm)

10

Involved

Lumpectomy + ALND

Yes (IDC)

IDC

Opacity (7 mm)

12

Involved

Lumpectomy + ALND

Yes (IDC)

ILC

Opacity (10 mm)

18

Involved

Lumpectomy + ALND

Yes (ILC)

IDC + IC

Scattered micro + opacity

12

Involved

Mastectomy + IBR

Yes (IC)

IDC + IC

Scattered micro + opacity

15

Involved

Lumpectomy + ALND

Yes (IC)

IDC + IC

Zone of micro (5 mm)

5

Involved

Lumpectomy + ALND

No

IDC + IC

Zone of micro (4 mm)

10

Involved

Mastectomy + ALND

Yes (IDC)

IDC + IC

Opacity (5 mm)

12

Involved

Lumpectomy + ALND

Yes (IC)

IC, ductal carcinoma-in-situ; micro, microcalcifications; IBR, immediate breast reconstruction; IDC, invasive ductal carcinoma; ALND, axillary lymph node dissection; ILC, invasive lobular carcinoma.

For the therapeutic evaluation, systematic lumpectomy and clips in the excision bed were performed at the end of the procedure. Histological results are listed in Table 3. In 41.1% of all cases, the SiteSelect biopsy margins were involved with tumor. The margins were involved in 33.3% of the focus lesions. With systematic lumpectomy at the end of the procedure, two involved margins were changed into a clear excision bed, and two second surgical procedures were avoided. The margins were clear for mammographic lesions <9 mm except for in two patients. A second operation under general anesthesia was not performed in 11 cases because of a complete resection (n = 5), old age (n = 3), associated morbidity (n = 2), and refusal of axillary dissection (n = 1).
TABLE 3

Invasive and noninvasive carcinomas (22-mm cannula)

Patient no

Histological type

Mammographic image

Size(mm)

SiteSelect margins

Margins after systematic lumpectomy

Surgery

1

IC

Zone of micro (6 mm)

3

Involved

Involved

Mastectomy + IBR

2

 

Zone of micro (7 mm)

50

Involved

Free

No

3

 

Zone of micro (10 mm)

4

Free

Free

No

4

 

Opacity (15 mm)

15

Free

Free

No

5

IDC

Opacity (4 mm)

6

Free

Free

No

6

 

Opacity (6 mm)

5

Free

Free

ALND

7

 

Opacity (6 mm)

6

Free

Free

ALND

8

 

Opacity (7 mm)

7

Free

Free

No

9

 

Opacity (7 mm)

15

Free

Free

No

10

 

Opacity (8 mm)

4

Free

Free

No

11

 

Zone of micro (5 mm)

5

Free

Free

No

12

 

Zone of micro (5 mm)

20

Involved

Involved

Lumpectomy + ALND

13

 

Scattered micro

25

Involved

Involved

No

14

ILC

Opacity (7 mm)

8

Free

Free

No

15

IDC + IC

Scattered micro

10

Involved

Involved

Mastectomy + ALND

16

 

Zone of micro (9 mm)

18

Involved

Involved

Lumpectomy + ALND

17

 

Opacity (10 mm)

13

Involved

Free

No

IC, ductal carcinoma-in-situ; micro, microcalcifications; IBR, immediate breast reconstruction; IDC, invasive ductal carcinoma; ALND, axillary lymph node dissection; ILC, invasive lobular carcinoma.

One third of the patients were reviewed at 6 months to evaluate the cosmetic results. The appearance of the breast was satisfactory in all cases and was good to excellent in 90% of cases. No subcutaneous depression was observed. Systematic mammography was performed at 6 months in all patients. It was normal in every case (except for the two biopsy failures). Lumpectomy was performed in one of these cases and revealed fibrocystic disease. The second patient was followed up, and the image was stable after 6 months. Ninety-eight percent of patients were very satisfied with the procedure. The remaining 2% were the patients who experienced pain during the procedure.

DISCUSSION

With the extension of mammographic screening, more and more breast cancers are discovered at a subclinical stage, and most measure <1 cm in diameter. At the Institut Curie, where we treat >1500 new breast cancers per year, 30% are nonpalpable, and this rate will continue to increase. We were, therefore, motivated to undertake a prospective study to assess a new radioguided stereotactic excision device to define its feasibility, complications, and indications. Two stereotactic surgery systems are currently available: the ABBI system (marketing stopped in September 2002) and, more recently, the SiteSelect system. These techniques allow en-bloc excision of breast tissue under local anesthesia and are based on identical principles, but they differ in terms of superficial tissue resection. The ABBI system has been predominantly analyzed in the literature.519 Our study is the most important prospective study evaluating the SiteSelect device.

The advantage of radioguided surgery over the standard vacuum-assisted core biopsy, Mammotome, is that it ensures en-bloc resection of the lesion and the adjacent breast tissue, thus allowing reliable histological examination with no risk of underestimation.5,6,20 The size of the cylinder excised is 15 or 22 mm wide and 50 mm long. One could thus imagine that such devices could be used to treat small subclinical breast cancers and that they might therefore constitute both a diagnostic and a therapeutic procedure.

In the preliminary study, we deliberately limited ourselves to excision of ACR stage III and ACR stage IV mammographic lesions. The SiteSelect achieved resection of the lesion in 96.9% of cases, and this is equivalent to the rate published with the first SiteSelect study.21 Combined with an immediate complementary surgical procedure under local anesthesia, resection was achieved in 98.7% of cases. This immediate surgery should be considered an integral part of the procedure, and we advocate that surgeons performing radioguided surgery be ready to convert to such a complementary excision when needed, thus improving the success rate of the procedure. The rate of complete excision with SiteSelect is similar to the rate of classical lumpectomy at the Institut Curie. Lumpectomy can be difficult for small mammographic lesions, particularly deep lesions with significant breast volume. In these cases, lumpectomy is generally performed under general anesthesia. The SiteSelect procedure is performed with the patient under local anesthesia and can be a good alternative to classic lumpectomy. With regard to tolerance and complications, the SiteSelect is a very safe tool: the only complication observed was an intramuscular hematoma that was drained immediately. This complication should have been avoided, because the microcalcifications were located adjacent to the pectoralis muscle, in which case the procedure should have been contraindicated. In all the other procedures, the biopsy was very well tolerated, with no difference from a standard 11-gauge vacuum-assisted core biopsy. The mean operative time was 42 minutes, and 98% of patients were very satisfied with the procedure. The mean operative time is 30 minutes for a classic lumpectomy.

One of the most frequent criticisms of the ABBI system was poor cosmetic results with no benefit over percutaneous biopsies or traditional surgery.22,23 The SiteSelect allows targeted resection of the lesion and adjacent breast tissue without resecting the subcutaneous tissue and could therefore reduce some of the poor results associated with the ABBI system22,23 by preserving superficial breast tissue that is sutured back after the procedure. We systematically evaluated the cosmetic results after SiteSelect excision, analyzing the location and the appearance of the scar. Before each procedure, we delimited a “no-scar zone,” including the neckline zone, in which a scar would be esthetically unsightly. The incision was deliberately performed in the lateral quadrants, even when the lesion was located in the medial quadrants: only 12.6% of scars were located medially. Late cosmetic results were excellent: no skin retraction or breast defects were observed, and the appearance of the scar was always excellent, never measuring >2 cm for the 15-mm cannula and 2.5 mm for the 22-mm cannula. Our data demonstrate that this technique can therefore be used to perform en-bloc resection of a mammographic abnormality in >96.9% of cases, with minimal cosmetic sequelae and excellent patient tolerance.

The second step of the study considered the 22-mm cannula as a therapeutic procedure to resect small breast cancers. Minimally invasive surgical techniques are currently being investigated to try to avoid traditional surgery for some patients with small, nonpalpable breast cancers. This constitutes a promising approach, and many different techniques are currently under evaluation. They comprise in situ destruction by cryoablation, thermoablation, or radiofrequency.24,25 The major disadvantage of these techniques is that they induce tumor destruction but do not resect any breast tissue. They therefore do not allow pathologic examination of the specimen, and this makes it difficult to demonstrate total tumor destruction. Radioguided surgery is another option, because it allows resection of the tumor and the surrounding breast tissue. However, the amount of tissue that can be excised is limited by the diameter of the cannula (15 or 22 mm for the SiteSelect). For therapeutic purposes, one should be able to ensure complete excision of the tumor and to obtain free margins of the specimen. In the first step of our study, the margins of the biopsy specimen were involved in 86.7% of cases with the 15-mm cannula, an unacceptable rate if one were to propose this procedure for treatment. For that reason, we think that this cannula is not adapted for a therapeutic procedure, and we decided to develop the 22-mm-diameter cannula. Our rate of margin positivity is equivalent to that published with the ABBI system and Corn’s study21 for the 15-mm cannula, in which, when specified, the positive margins rates varied from 45% to 93%.519 As defined in our protocol, all patients with in situ or invasive carcinoma (22 cases) underwent wide excision of the biopsy cavity: residual tumor was present in 64.2% of cases. This protocol was not specified for the first prospective SiteSelect study,21 even for clear margins, as in our study. Of note was the fact that no residual tumor was observed in patients with uninvolved margins on the initial biopsy, thus demonstrating that complete tumor resection can be achieved by radioguided surgery.

To our knowledge, only two other authors have performed systematic re-excision of the tumor bed after radioguided surgery.13,18 Lifrange and Colin13 performed 36 ABBI procedures: 18 patients had a carcinoma, and in 11 (61%) of these cases, the biopsy margins were involved with tumor. Surgical re-excision was performed in all 18 cases; 63.6% of patients had residual tumor. Damascelli et al.18 performed 59 ABBI procedures, and malignant tumor was detected in 25 cases. The biopsy margins were involved with tumor in 72% of these cases, and residual tumor was present on the complementary lumpectomy specimen in 77% of cases. Some authors have suggested that the high positive margin rate with radioguided surgery is due to the surgeon’s inexperience in the technique, and there is a learning curve, after which the rate of margin involvement should be reduced.16 However, in our study, as in those of Lifrange and Colin13 and Damascelli et al.,18 all procedures were performed by the same surgeon. In our own experience, the rate of margin involvement was identical throughout the series. This hypothesis therefore seems unlikely.

It is more likely that the main limiting factor is the size of the lesion compared with the diameter of the cannula, because the mammographic image can sometimes underestimate the histological size, particularly in the case of intraductal carcinoma. The 15-mm cannula therefore seems too narrow for tumor excision.

For the 22-mm cannula, the margins of the biopsy are involved in 41.1% of cases. If we exclude scattered microcalcifications for which the procedure is merely diagnostic, this rate decreases to 33.3%. With the association of systematic lumpectomy at the end of the procedure, a clear excision bed is obtained in 80% of the cases. For all the lesions with a mammographic size of <9 mm, the margins are clear. A second operation under general anesthesia was not performed in 11 cases because of complete excision (patients 2, 3, 4, 10, and 11), old age (patients 5, 8, and 14), comorbidity (patients 9 and 13), and refusal of axillary dissection (patient 17). Corn’s study21 was too small (n = 125) to analyze the 22-mm cannula because only three 22-mm procedures were performed, with two clear margins.

The 22-mm cannula should therefore constitute a better solution in treating small invasive carcinoma, particularly in older women and in patients with high comorbidity, because of local anesthesia, a mean procedure time similar to that for lumpectomy, and a high success rate. However, the indications must be restricted because of the high rate of involved margins compared with classic lumpectomy. These indications are actually mammographic ACR stage IV or ACR stage V lesion (mass or microcalcifications) with a mammographic size <9 mm. We have now started a new study with this wider cannula and have modified our protocol to achieve complete en-bloc excision of subcentimeter breast carcinomas, with free margins to associate, in the future, axillary sentinel node dissection with the SiteSelect procedure.

CONCLUSION

Radioguided stereotactic surgery allows en-bloc resection of subclinical breast lesions without fragmentation. This procedure allows resection of the lesion under local anesthesia in 98.6% of cases with excellent patient satisfaction. SiteSelect can thus be used to resect undetermined lesions, with excellent long-term cosmetic results. The 15-mm cannula is not adapted for a diagnostic procedure compared with vacuum-assisted biopsies and is not adapted for therapeutic procedures because of the low rate of clear margins. Preliminary results with the 22-mm cannula show that it is possible to overcome this obstacle. We have now started a second study with this cannula and hope to be able to demonstrate that radioguided surgery is a possible tool for minimal invasive surgery for subcentimeter breast cancers, with restricted indications, particularly for older women or patients with high comorbidity.

Copyright information

© The Society of Surgical Oncology, Inc. 2005