Twenty two patients that were underwent EA-BCS for breast cancer at the Department of Breast Care Center, Myongji Hospital, Kwandong University College of Medicine from June 2006 to February 2008. These patients were selected for EA-BCS according to the following criteria: 1) early breast cancer without any severe co-morbid conditions such as heart disease, liver dysfunction, renal failure, or a poor performance status, 2) no tumor extension to the nipple or direct invasion to the skin and pectoralis muscles, 3) no axillary lymph node metastasis, 4) no multifocal lesions and 5) no microcalcified lesions which could not localize by preoperative ultrasonography. The patients were evaluated by preoperative studies such as mammography, ultrasonography, chest computed tomography (CT), and magnetic resonance imaging (MRI). Informed consent was obtained from all the patients prior to performing EA-BCS.

In addition to the basic endoscopic monitoring system (Stryker®; Stryker Endoscopy, San Jose, USA), Figure 1 shows the special tools we used for this surgery: 1) an Alexis® Wound Retractor (2-4 cm, XS; Applied Medical, Rancho Santa Margarita, USA) (Figure 1A) was used for axillary incision wound protection 2) Bipolar scissors BiSect (ERBE, Tuebingen, Germany) (Figure 1B) and a Harmonic Scalpel™ (Ethicon Endo-Surgery, Cincinnati, USA) were used for accurate resection and hemostasis 3) a Visiport™ Plus (5-11 mm; Tyco Healthcare, Norwalk, USA) (Figure 1C) was used for sentinel lymph node biopsy (SLNB) and axillary lymph node dissection (ALND) 4) an ENDOPATH® (Ethicon Endo-Surgery) (Figure 1D) was used for penetration of subcutaneous tissue, 5) an Endosector LE (Curexo, Anyang, Korea) (Figure 1E) and an Ultra Retractor (Johnson & Johnson KK, Tokyo, Japan) (Figure 1F) were used for dissection of the retromammary space. The endoscope is a rigid, straight instrument that is 5-mm in diameter, it is oblique at 30° and it is 10-mm in diameter and flat at 0°.

Under general anesthesia, the patient was placed in the supine position with the ipsilateral arm abducted to 90° without disturbing the operative field. The breast, chest wall, axilla, and ipsilateral arm were prepared and draped in the usual sterile fashion. The location of the tumor was confirmed by means of palpation and performing intraoperative ultrasonography, even if the tumor was non-palpable lesion. The tumor location and the quadrant to be excised were marked on the skin. A SLNB was performed with the dye-staining method or radioisotope-guided method, or a combined method at the beginning of the operation before gland resection on a case by case basis. For the cases using the dye staining method, we subcutaneously injected 5 mL of 0.8% indigocarmine in the periareolar region, and we performed breast massage for 10 min. A 3.5-cm skin incision was made along the line of the axillary skin crease. A Visiport™ Plus and 10-mm/0° scope was inserted from the axillary incision, and the stained lymph nodes were found with using endoscopy assistance. In the patients who were found to have a positive sentinel lymph node, an ALND was carried out at levels I and II through the same incision under endoscopy assistance. After completing the SLNB, the Alexis® Wound Retractor was inserted into the incision. Through the axillary incision, we cut the adipose tissue deep into the shallow lateral chest fascia, and we slide it to the lateral edge of the major pectoral muscle; we then bluntly dissected the fascia beyond the retromammary space of the quadrant where the tumor was to be excised by using an Endosector LE or an Ultra Retractor under 5-mm/30° scope assistance. The fiberoptic breast retractor and a 10-mm/0° scope were inserted into the subpectoral pocket, and further dissection was performed using the Bipolar scissors BiSect or Harmonic Scalpel™ under endoscopic guidance. After completing the dissection of the retromammary space, we went through the subcutaneous tissue. A skin flap was made by the tunnel method.(10) Further, we made a 3-cm periareolar skin incision in the direction of the tumor. The ENDOPATH® which has an opening that accommodates a 10-mm/0° scope was next inserted beneath the skin flap of the periareolar skin incision. The skin flap was made by blunt thrusting of the ENDOPATH®. We completed the subcutaneous flaps of the tumor quadrant with using the Bipolar scissors BiSect or the Harmonic Scalpel™. The Bipolar scissors BiSect or Harmonic Scalpel™ were also used to vertically dissect the mammary gland. We partially removed the mammary gland with free surgical margins at least 1 cm away from the tumor border. The resected gland was pulled out through the axillary Alexis® Wound Retractor port. We stained the superficial cut surface of the resected gland with India ink, with placing one or more sutures for directional orientation. The resected gland was brought to the pathology department and the frozen sections of its cut margin were immediately examined by a pathologist. If the cut margin was positive, then additional resection was required until a negative cut margin was confirmed. After the resection was completed, meticulous dissection of the residual tissue and hemostasis could be achieved with using the Harmonic Scalpel™ and an electrocoagulator under endoscopic guidance. After saline irrigation, the remaining mammary glands were mobilized and sutured together to preserve the original breast shape. A suction drainage was left in place before closing the surgical wound. We reconstructed the breast by immediate volume replacement as needed. We placed an absorbable implant into the dead space. The absorbable implant was made of Vicryl mesh® (polyglactin 910 Mesh; Ethicon, Johnson & Johnson, Somerville, USA) wrapped with Interceed® (Oxidized regenerated cellulose; Ethicon, Johnson & Johnson).(11) In cases of immediate volume replacement, however, we did not use a suction drainage to preserve breast contour by seroma formation. The surgical wound was closed with a subcutaneous reversed suture using 4-0 Vicryl. Compression was provided with a surgi-bra to prevent postoperative bleeding and hematoma formation.

All the patients were regularly examined and questioned about their satisfaction with the aesthetic results every six months follow-up. Many methods of cosmetic assessment have been reported.(12-14) We used the 5-item-4-step method (ABNSW) designed by Yamashita.(12) The method of ABNSW is the modified method of the MDACS (malposition, distortion, asymmetry, contour deformity and scar) grading system.(13) ABNSW contains 5 items: asymmetry (A), breast shape (B), nipple deformation (N), skin condition (S) and wound scar (W). All items can be scored by the patients themselves under the 4 steps of the grading system as follows: 0: poor, 1: fair, 2: good, 3: excellent. The scores for the 5 items were then combined to obtain a total score. On a scale of 15 points, the summed points were defined as excellent: 15 points, good: 11-14 points, fair: 6-10 point and poor: <5 points. The patients were also requested to evaluate the subjective satisfaction of their postoperative aesthetic results as poor, fair, good, or excellent using 4-step method of ABNSW at every six months after surgery.

The clinical/pathological features of all the patients are presented in Table 1. The average age was 52.0 yr (range, 36-74 yr). The locations of the tumors are summarized in Table 1. The upper-outer quadrant was the most common site (50.0%). The mean tumor size was 2.2 cm (range, 0.7-5.5). TNM staging was done according to the American Joint Committee on Cancer staging system. All the patients were negative for lymph node metastasis on the preoperative evaluation, and this included ultrasonography or MRI. A SLNB was performed in all patients. Four patients who had positive sentinel lymph node frozen biopsy results underwent ALND, and only one out of four patients had positive axillary node biopsy results at level I and II. For the postoperative diagnosis, 81.8% (18 out of 22) of the patients were negative for metastatic lymph nodes; however, 18.2% (4 out of 22) of the patients were positive for lymph node metastasis; distant metastasis was not observed in any patient. The disease in all the patients with malignancies was classified as earlier than stage IIB. Estrogen receptor positive results were obtained in 86.4% (19 out of 22) of the patients and the progesterone receptor status was positive in 95.5% (21 out of 22) of the patients. We defined more than three scores at the sum of proportional and intensity score as hormone receptor positive. An over-expression of HER2/neu was observed in 40.9% (9 out of 22) of the patients. The histopathological diagnoses of treated diseases are summarized in Table 1.

EA-BCS was performed for 22 patients with breast cancer. A SLNB was performed in 22 patients. An ALND was performed in four patients who had positive sentinel lymph node frozen biopsy results. Immediate volume replacement treatment was carried out in 11 patients with an absorbable implant that was made of Vicryl mesh® wrapped with Interceed®. The mean operative time was 144 min (range, 105-190 min). The mean volume of the extracted mass was 219.2 cm³ (range, 56-1332 cm³).

Ninety five percent (21 out of 22) of the patients were questioned about their satisfaction with the aesthetic results every six months follow-up. There was one patient who underwent a skin sparing mastectomy (SSM) with immediate breast reconstruction due to a positive resection margin of the nipple specimen on the permanent report was excluded in this assessment. The postoperative evaluations after the procedures are shown in Figure 2. The scar from the wound was inconspicuous. The shape of the breast appeared natural and nearly symmetrical. The patients were satisfied with the operation. The aesthetic results were evaluated using the original scoring system ABNSW.(12) The mean total ABNSW score was 12.2 (Table 2). By comparing with 4-step method, scores greater than 11 points were considered good or excellent; subsequently, 95% of the evaluated cases had good or excellent results. In addition, the degree of satisfaction was evaluated using 4-step method of ABNSW. 100% (21 out of 21) of the patients scored good or excellent results. Almost all the patients were satisfied with the aesthetic results of their surgery.

Table 3 summarizes the surgical complications. The postoperative complications were as follows: lymphedema developed in 9.1% (2 out of 22), skin retraction developed in 9.1% (2 out of 22) and wound infection developed in 13.6% (3 out of 22). Two out of these 3 patients had wound infection because of the prosthetic infection. All cases of wound infection were developed in diabetes patients. Skin retraction was caused by reoperation of infected prosthesis removal. Three patients underwent reoperation. There were 9.1% (2 out of 22) of the patients who had to have their prosthesis removed because of infection of the Vicryl mesh®. We did not additional procedure to preserve breast shape after the infected prosthesis removal. In addition, 4.5% (1 out of 22) of the patients had a positive resection margin of the nipple specimen on the permanent pathology report. Therefore, we performed a SSM with immediate breast reconstruction nine days after the initial operation. To detect local recurrence and distant metastasis, after surgery, ultrasonography was performed every six months and positron emission tomography-CT was performed one year after the operation. The patients underwent standard adjuvant therapy based on the recommendation of St. Gallen.(15) After a mean follow-up period of 24.0 months (range 12-32 months), neither local/regional recurrence nor distant metastases were detected.