| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Departments of Surgery (CN, SG, RJS, KBC), Pathology (PF), and Biostatistics (MCF), Institut Curie, Paris, France.
Correspondence: Address correspondence and reprint requests to: Dr Claude Nos, Department of Surgery, Institut Curie, 26 Rue dUlm, 75005 Paris, France; Fax: 33-144324006.
| ABSTRACT |
|---|
|
|
|---|
Methods: Two hundred fifty-three women consecutively identified with operable breast cancer underwent sentinel lymph node (SLN) detection by the patent blue dye method followed by completion axillary lymph node dissection. A standard pathological examination was performed for all SLN. Then, a pathological color quality assessment (PCQA), which checked for the presence of the blue dye, was performed on the paraffin blocks of the nonmetastatic SLN. Six preoperative identifiable variables likely to influence the detection rate were examined.
Results: The surgical detection (sd) rate was 84% (213 of 253) and the PCQA rate was 73% (185 of 253). Only breast size (sd, P = .0005; PCQA, P = .0007) and body mass index
30 (sd, P = .005; PCQA, P = .0007) were significant for SLN identification. Multivariate analysis revealed two independent factors influencing SLN identification: breast size (sd, P = .0001; PCQA, P = .002) and the timing of injectioninjection prior to lumpectomy (sd, P =.04).
Conclusions: The optimal patient features for identifying the SLN by the patent blue dye method are small or medium-sized breasts, low body fat, and that the procedure is carried out prior to tumor excision. The PCQA offers a useful second assessment of the surgically removed SLN, introducing an independent element of quality control.
Key Words: Sentinel lymph node biopsy Breast carcinoma Axillary dissection Pathological color quality assessment
| INTRODUCTION |
|---|
|
|
|---|
| PATIENTS AND METHODS |
|---|
|
|
|---|
Six variables which might influence the SLN identification were tested: surgeons experience, patients age, body mass index (BMI), timing of injection in relation to tumor excision, tumor location, and breast size.
Surgeons experience:
The first 10 cases performed by each surgeon (one surgeon only performed 8 cases) were identified separately (Table 1). This group of 68 cases was then compared with the remaining 185 patients (Table 2).
|
|
50 years (62 patients) and >50 years (191 patients).
Body mass index (BMI):
This was calculated using the Quetelet index: weight (Kg)/height2 (M). BMI information was available on only 234 patients. Twenty-five patients with a BMI >30 made up the obese group and 209 patients with a BMI
30 made up the nonobese group.
Timing of injection in relation to tumor excision:
All patients were injected with 2 to 5 ml of patent blue dye. The injection was performed intraglandularly and subcutaneously but not intradermally. Breast massage was performed systematically. The injection was peritumoral and before lumpectomy or mastectomy in 165 patients for whom malignancy had been confirmed by fine needle aspiration cytology or core needle biopsy. In 88 patients, the blue dye injection was performed around a biopsy cavity. This was either after frozen section had been used to confirm the malignant nature of the lesion (49 patients) or when an excisional biopsy had been performed as a previous operation (39 patients).
Tumor location:
This was located in the medial (57 patients) or lateral (196 patients) hemisphere of the breast. The hemispheres were defined by using a vertical line passing through the medial border of the areola. Central or subareolar lesions (25 patients) were considered to be in the lateral hemisphere.
Breast size:
Breast size information was available on only 237 patients. Patients were divided into two groups according to bra cup size: cup size D-E (73 patients had large breasts) and cup size A-B-C (164 patients had small or average breasts).
The SLN identified by the surgeon determined the surgical detection rate (i.e., whether one or more SLN were identified for a given patient). The SLN biopsy and completion axillary lymph node dissection were sent to the pathologists. No frozen section was performed on the SLN biopsy. The SLN biopsy and all axillary lymph nodes were bisected and fixed in AFA (alcohol, formalin, and acetic acid) in the following proportions: 5% of 100%-acetic acid, 75% pure ethyl alcohol, 18% demineralized water, and 2% of 40%-formalin. Then they were embedded in paraffin individually. One level of hematoxylin- and eosin-stained slide from each node was examined for the presence of metastatic disease. Then, the nonmetastatic SLN underwent a pathological color quality assessment (PCQA) for the presence of dye. The pathologist performed this by checking the paraffin blocks macroscopically and assessing whether at least one SLN identified by the surgeon was blue or not for any given patient.20
The PCQA rate was then defined as the sum of the nonmetastatic SLNs which were blue stained (as confirmed by the pathologist) and the total number of metastatic SLNs. This was then divided by the total number of patients to determine the PCQA rate, which was expressed as a percentage.
PCQA rate = PCQA positive nonmetastatic SLN + metastatic SLN/total number of patients.
The surgical detection rate and PCQA rate were examined in relation to the different variables previously identified. A multivariate analysis was then performed, using the logistic regression model.21 A forward stepwise procedure was used to select the variables entered in the model. Odds ratios [OR] are presented with their 95% confidence interval [CI].
| RESULTS |
|---|
|
|
|---|
The surgical detection rate of SLN was 84.2% (213 of 253). After standard pathological examination, 61 SLN were found to be metastatic (29%, 61 of 253), 152 SLN were free of disease (71%, 152 of 253), of which 6 SLN were false negatives. The false negative rate was thus 9% (6 of 67). The negative predictive value was 96%. Half the false negative cases arose during the first 10 cases for each surgeon (Table 1). All the false negative cases occurred for tumors located in the lateral hemisphere. One false negative occurred in a patient after prior excisional biopsy.
Univariate analysis found the two factors that influenced the detection of SLN to be breast size and BMI. Patients with small or normal-sized breasts had a better surgical detection rate than patients with large breasts (P = .0005). Nonobese patients had a better surgical detection rate than patients in the obese group (P = .005). Other factors were not significant (Table 2). Multivariate analysis found that two factors remained significant for SLN surgical detection: (a) breast size (P = .0001) and (b) timing of injection in relation to tumor excisioninjection prior to surgery compared to injection after lumpectomy (P = .04) (Table 3).
|
|
| DISCUSSION |
|---|
|
|
|---|
We used the peritumoral injection method which is the most widely reported in the literature.27,9,10,1219
The SLN was detected by following the blue lymphatic pathway(s) to its first node in the axilla. So, in principle, the tumor, the lymphatic vessel, and the SLN should be stained blue because there is a continuum of the metastatic lymphatic diffusion.
The determining factors for the successful application of the blue dye method alone in identifying sentinel lymph nodes are not clearly established. In our series, we used the surgical detection rate, PCQA rate, and the false negative rate to determine which patients would obtain a satisfactory result from the use of the blue dye method and would not require radioactive colloid detection. This was defined in terms of correctly predicting the axillary status. It is generally agreed that approximately 30 cases of SLN biopsy and completion axillary lymph node dissection are required for a surgeon to become proficient in the SLN technique, and the surgeons experience is the best known predictive factor for obtaining a reliable result.15 In our series, surprisingly, surgical experience was not a determining factor in identifying the SLN. This held true for both the surgical detection rate and the PCQA rate. We considered that the first 10 cases would constitute the surgeons learning curve, taking into account that all seven surgeons worked closely as a team and could benefit from each others experience.3,17 However 3 of 6 false negative results occurred during the surgeons first 10 cases (Table 1). In this series, the learning period did not appear to influence the identification of the sentinel node, but it is important for minimizing the false negative rate.
Univariate analysis of the surgical detection rate and PCQA identification rate shows that the predictive factors for identification of the SLN by blue dye are the same: namely, breast size and BMI. A multivariate analysis of surgical detection rate and PCQA identification rate shows that breast size is the most significant factor in identifying the SLN.
Interpretation of breast size is, however, subjective and has rarely been analyzed in other articles. The identification rate is almost certainly linked to the size of the axilla. The search for blue-stained lymphatic vessels and blue SLN is more difficult in a voluminous and adipose axilla than if the patient is thin. The body mass index is an independent factor in this study. In obese patients, the breasts are fatty and the capacity of lymph nodes to retain the blue dye may be decreased because lymph nodes can be replaced by fat in these patients. Age (over 50 years) is also a factor that impacts the identification rate of the sentinel node10 even though it was not found to be statistically significant in this study. Krag10 showed, using the isotope technique alone, that age over 50 was an independent factor in lowering the identification rate of the SLN, possibly due to lymph node fatty replacement. Therefore, it would seem that the search for the SLN in obese patients is more difficult not only by the blue dye method but also by the isotope method. Under these conditions, in order to increase the identification rate for these patients, it is probably necessary to use both the isotope and blue dye techniques. Increasing radioactive tracer doses may also be helpful.10
The second independent factor influencing SLN identification was the timing of injection in relation to the lumpectomy. SLN biopsy should be performed prior to tumor excision. This parameter is significant in surgical identification of the SLN but is linked to age for identification by PCQA. Thus, peritumoral injection is particularly beneficial to young patients because it increases the chances of finding a blue node. Injection in the tumor bed, after excision for frozen section confirmation of malignancy or prior excisional biopsy, leads to significantly lower SLN identification rates.
Opinions in literature differ concerning the identification rate of the SLNby injection of the dye or radioactive tracer in the tumor bedafter prior excisional biopsy. Some investigators have observed lower identification rate of SLN after prior excisional biopsy when using vital blue dye (as in our study) or radioactive colloid.10,12 Others have had equal success regardless of a prior biopsy, both using vital dye6,7 or combined techniques.16,18 However, these same investigators do not advise using SLN biopsy for patients with extensive previous surgical biopsy (specimen >6 cm),6 especially when located in the upper outer quadrant of the breast.16 In the Memorial Sloan-Kettering16 series, most of the false negative cases occurred when injection was performed into the tumor bed after excisional biopsy.
A prior excisional biopsy therefore constitutes a relative contraindication for the SLN technique that uses a peritumoral injection. Consequently, after analyzing the results of our study, we now recommend the search of the SLN by peritumoral injection after preoperative diagnosis by fine needle aspiration or core needle biopsy. Where tumors have already been removed by large excisional biopsies, alternative methods of injection such as subareolar may prove more reliable, but further studies are required. Lastly, medial location of the tumor decreased, but not significantly, the blue dye identification rate of the SLN in the axilla. This is at least partially due to the physiologic drainage of the breast, because not all the areas of the breast drain towards the axilla.13 The absence of a SLN in the axilla is not necessarily a failure of the technique. It should however, ideally, lead to further inspection for a SLN in other nodal fields (i.e., internal mammary, infra/supra clavicular). This is unfortunately not possible with the blue dye method alone, and it, therefore, justifies isotope mapping with colloid injection using a peritumoral radioactive colloid injection in order to reveal nonaxillary lymph drainage. However, it is noteworthy that the accuracy of the axillary SLN, in terms of false negative rate for medially located tumors, appears good. In our series, no false negative were found for a medially located tumor. This was also the finding of the Memorial Sloan-Kettering16 series and in the multicenter study carried out by Krag et al.10 However, both studies used isotope detection.
Regarding our study, breast carcinoma patients who had a combination of two criteria: small to medium-sized breasts and no excision biopsy of the tumor (113 patients), had a surgical detection rate of 93.5% and PCQA rate of 87%. On the other hand, the 15 patients of this current series with large-sized breasts, and who had undergone tumor excision prior to SLN biopsy, had a surgical detection rate which was only 53% and PCQA rate of 46%.
| CONCLUSION |
|---|
|
|
|---|
| Acknowledgments |
|---|
Received for publication September 8, 2000. Accepted for publication January 25, 2001.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. Nakahara, Y. Kitagawa, H. Yakeuchi, H. Fujii, T. Suzuki, M. Mukai, M. Kitajima, and A. Kubo Preoperative Lymphoscintigraphy for Detection of Sentinel Lymph Node in Patients with Gastric Cancer--Initial Experience Ann. Surg. Oncol., May 1, 2008; 15(5): 1447 - 1453. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Alran, Y. De Rycke, V. Fourchotte, H. Charitansky, F. Laki, M. C. Falcou, M. Benamor, P. Freneaux, R. J. Salmon, for the Institut Curie Breast Cancer Study Group, et al. Validation and Limitations of Use of a Breast Cancer Nomogram Predicting the Likelihood of Non Sentinel Node Involvement After Positive Sentinel Node Biopsy Ann. Surg. Oncol., August 1, 2007; 14(8): 2195 - 2201. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Lerman, G. Lievshitz, O. Zak, U. Metser, S. Schneebaum, and E. Even-Sapir Improved Sentinel Node Identification by SPECT/CT in Overweight Patients with Breast Cancer J. Nucl. Med., February 1, 2007; 48(2): 201 - 206. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Louis-Sylvestre, K. Clough, B. Asselain, J. R. Vilcoq, R. J. Salmon, F. Campana, and A. Fourquet Axillary Treatment in Conservative Management of Operable Breast Cancer: Dissection or Radiotherapy? Results of a Randomized Study With 15 Years of Follow-Up J. Clin. Oncol., January 1, 2004; 22(1): 97 - 101. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Nos, P. Freneaux, C. Louis-Sylvestre, J. S. Hurren, D. Heitz, X. Sastre-Garau, and K. B. Clough Macroscopic Quality Control Improves the Reliability of Blue Dye-Only Sentinel Lymph Node Biopsy in Breast Cancer Ann. Surg. Oncol., June 1, 2003; 10(5): 525 - 530. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. A. Litvak and A. E. Giuliano The Results of the Department of Defense's Breast Lymphatic Mapping Study Ann. Surg. Oncol., June 1, 2002; 9(5): 428 - 429. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |