The American Society of Clinical Oncology (ASCO) published guidelines for lymphatic mapping and SLND [4]. Routine SLND is acceptable for early-stage breast cancer in patients with clinically node-negative disease. The guidelines are summarized in Table 1. There is a debate on the timing of SLND in patients undergoing neoadjuvant chemotherapy (NAC). According to the ASCO guidelines, SLND is considered an option before, but not after, preoperative chemotherapy. However, practice patterns differ, with many clinicians performing SLND before or after NAC. Data can support either option. The National Cancer Institute Conference concluded that SLND can be performed before or after preoperative chemotherapy in patients with clinically node-negative cancer [5].

Table 1 ASCO guidelines 2005

Abbreviations: DCIS, ductal carcinoma-in situ; SNB, sentinel lymph node biopsy.

a Levels of evidence: Good, multiple studies of SNB test performance based on findings on completion ALND; limited, few studies of SNB test performance based on findings on completion ALND or multiple studies of mapping success without test performance assessed; and insufficient, no studies of SNB test performance based on findings on completion ALND and few if any studies of mapping success.

From Lyman G, Giuliano AE, Somerfield M, et al. American Society of Clinical Oncology Guideline recommendations for sentinel lymph node biopsy in early-stage breast cancer. J Clin Oncol 2005;23:7703–20; with permission.

Lymphatic mapping may be achieved with vital blue dye and/or radioactive tracer. After induction of general anesthesia, the breast, chest wall, axilla, and arm are prepared and draped. An injection of 3 to 5 mL of isosulfan or methylene blue dye is given into the breast parenchyma on the axillary side of the breast mass. If the tumor is nonpalpable, needle wire localization or ultrasonography may be used as a landmark for the dye injection. Alternatively, subareolar injection may be used. If the tumor was previously excised, the dye may be injected into the biopsy cavity wall but not into the cavity itself. The injected area is massaged to encourage lymphatic flow. Five minutes after the dye injection, a transverse incision is made in the axilla, just below the hair-bearing area. To facilitate exposure, the arm is brought over the head. With blunt dissection, the dye-filled lymphatic tract is identified and followed in the subfascial plane until a blue-stained sentinel node is found. Each dye-filled lymphatic tract should be followed, and all blue-stained sentinel nodes should be excised. For tumors located in the medial quadrant or the inner upper quadrant, preoperative lymphoscintigraphy may be helpful because these tumors may drain only into the internal mammary nodes or level III nodes, respectively.

Radioactive tracer requires a nuclear medicine team and preoperative injection and is the most commonly used technique. Technetium Tc 99m–labeled sulfur colloid [6] or albumin colloid [7,8] is injected in peritumoral, intradermal, or subareolar location 2 to 24 hours before surgery. The axillary drainage pattern is visualized on a lymphoscintigram or detected intraoperatively with a gamma counter. The skin incision is made, and in vivo radioactivity of the axillary nodal basin is reevaluated. A radioactive node is variably defined as the hottest node by absolute counts, a 10:1 ratio of sentinel node to background, a 4-fold reduction in count after sentinel node removal, or a 10-second count greater than 25 [6–10]. Any suspicious nodes that did not take up the dye or colloid should also be removed because cancer-filled nodes may not take up dye or colloid.

Grube and Giuliano [11] reviewed early published studies with 100 or more subjects reporting the identification and accuracy of SLND using vital dye, gamma probe–guided surgery or combined technique, followed by completion ALND for invasive breast cancer. The overall false-negative sentinel node identification rate ranged from 0% to 14.3%, and the accuracy ranged from 93% to 100%. Potential factors contributing to failure to identify sentinel node include variable lymphatic drainage pattern, patient or tumor characteristics, surgical technique, surgeon training, and pathologist experience.

The primary site of lymphatic drainage from the breast is the axilla. However, 1.3% to 9.9% isolated internal mammary drainage has been reported [8,12–15]. In a human cadaver study evaluating the lymphatic drainage of the breast [16], the lymphatics of the nipple-areola complex did not drain into the same nodes that receive the lymphatic drainage passing through the breast. At present, there is no consensus regarding the ideal sites for dye or radioactive colloid injection [17–19]. In practice, peritumoral [3,6], intradermal [7,20,21], or subareolar [22,23] injections seem equally effective to identify axillary nodes. However, peritumoral injections lead to better identification of the internal mammary nodes when compared with intradermal or subareolar injections [21,24–27]. Subareolar injection is regarded as a good choice for sentinel node identification for a multicentric disease [28].

Several studies have evaluated the proportion of successful SLND with blue dye, radiocolloid, or combination of dye with isotope with respect to sentinel node identification rate, accuracy, and/or false-negative rate. In the American College of Surgeons Oncology Group (ACOSOG) Z0010 trial, 5237 patients underwent SLND. Blue dye alone was used in 14.8%, radiocolloid alone in 5.7%, and the combination of dye with isotope in 79.4%. No statistically significant difference in sentinel node identification failure rate was seen (1.7%, 2.3%, and 1.2% for the blue dye, radiocolloid, and the combination of dye with isotope, respectively) [29]. Morrow and colleagues [30] performed a randomized trial comparing the use of blue dye alone with that of combined dye and isotope. The success rate of sentinel node biopsy was higher with combined mapping than with blue dye alone (100% vs 86%, P = .002). The accuracy and false-negative rates were similar. An accuracy of 100% for combined mapping versus 98% for blue dye and a false-negative rate of 0% for combined mapping versus 5% for blue dye were observed. In a study by Meyer-Rochow and colleagues [31], similar identification rates, accuracies, and sensitivities were reported with blue dye alone and triple modality, consisting of preoperative lymphoscintigraphy, intraoperative gamma probe, and intraoperative blue dye.

The accuracy of SLND depends greatly on the proficiency of the surgeon performing the procedure [32]. A learning curve exists, and surgeons master the procedure at different rates. According to the American Society of Breast Surgeons guidelines [33,34], 20 cases of SLND with back-up ALND with an identification rate of 85% and a false-negative rate less than 5% are recommended before abandoning ALND. However, the widespread use of this technique and its teaching in training programs have made these recommendations obsolete.

Studies have shown that increasing the number of sentinel nodes removed may increase accuracy and decrease false-negative rates [35–37]. In the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-32 trial, the false-negative rate was 17% for removal of 1 sentinel node, 10% for 2 removal of nodes, 6.9% for removal of 3 nodes, 5.5% for removal of 4 nodes, and 1% for removal of 5 or more nodes [35]. Zakaria and colleagues [36] evaluated how many sentinel nodes are enough in SLND for breast cancer and found that 98% of patients with lymph node metastasis were identified by the third node and 100% by the fourth node. They concluded that terminating the procedure at the fourth node may reduce the morbidity of the procedure. Chagpar and colleagues [37] evaluated whether removing 3 sentinel nodes is sufficient. When SLND was limited to the first 3 nodes, the false-negative rate was 10.3% in this University of Louisville Breast Cancer Sentinel Lymph Node Study. The investigators concluded that removal of only 3 nodes cannot be recommended. Controversy exists regarding how many nodes are sufficient for accurate staging of the axilla. In general, all blue nodes or nodes at the end of blue lymphatic channel, hot nodes or nodes with radioactive counts greater than 10% of the hottest node, and any palpably suspicious nodes should be removed. Increasing the number of nodes removed increases the probability of removing the sentinel node, but experienced surgeons may need to remove only 1 node for a successful sentinel node identification.

Several studies evaluated the patient and tumor characteristics associated with SLND success and failure. Age 70 years or older and increased body mass index (BMI), defined as the weight in kilograms divided by the height in meters squared, were associated with SLND failure in the ACOSOG Z0010 trial [29]. The differences in tumor location, the type of biopsy, and the number of SLNs removed significantly affected the false-negative rate in the NSABP B-32 trial [35]. The study result is summarized in Table 2. In the Axillary Mapping Against Nodal Axillary Clearance (ALMANAC) study, SLND success decreased with increasing BMI, tumor location other than the upper outer quadrant, and nonvisualization of hot nodes on the preoperative lymphoscintigraphy [38].

Several large randomized prospective trials have accumulated and evaluated the experience of sentinel node identification for staging. The goals of ALND are to maximize survival, provide regional control, and accurately stage the tumor. However, ALND is associated with significant morbidities including lymphedema, limited range of motion, pain, numbness, wound infection, seroma formation, and weakness of the arm.

The purpose of the NSABP B-32 trial is to establish whether SLND can achieve the same therapeutic goals as conventional ALND but with less morbidity [35]. A total of 5611 women with clinically node-negative operable breast cancer were randomized to SLND followed by ALND or to observation if the sentinel node was tumor free. The overall survival, disease-free survival, and regional control were statistically equivalent between the groups. When the sentinel node is histologically tumor free, SLND alone with no further ALND is an appropriate, safe, and effective therapy for patients with clinically node-negative breast cancer [39]. The morbidity results comparing SLND and ALND were recently reported [40]. The shoulder abduction deficit greater than 10% peaked at 1 week for the ALND (75%) and SLND (41%) groups. Arm volume difference of greater than 10% was seen at 36 months for the ALND (14%) and SLND (8%) groups. The rate of numbness peaked at 6 months for the ALND (49%) and SLND (15%) groups. These results indicate the superiority of SLND with respect to surgical morbidity outcomes over a 3-year follow-up period. The secondary end points were accuracy and technical success. A 97.2% sentinel node identification rate and 9.8% false-negative rate were reported. Only 1.4% of the sentinel nodes were found outside of the axillary levels I and II.

The ALMANAC trial quantified the detection and false-negative rates of SLND and evaluated the factors influencing them [38]