| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
SUPPLEMENT |
From the University Department of Surgery (DC, REM) and Department of Epidemiology, Statistics and Public Health (RGN), University of Wales College of Medicine, Heath Park, Cardiff, United Kingdom, on behalf of the ALMANAC Trialists Group.
Correspondence: Address correspondence and reprint requests to: Robert E. Mansel, MS, FRCS, Principal Investigator, ALMANAC Trial, University Department of Surgery, University of Wales College of Medicine, Heath Park, Cardiff CF14 4XN, United Kingdom; Fax: 029-2076-1623; E-mail: manselre{at}cf.ac.uk
ABSTRACT
Sentinel node biopsy (SNB) is a minimally invasive procedure to stage the axilla in patients with breast cancer. Like any new surgical procedure, it is associated with a learning curve. This article describes the learning curve as part of the ALMANAC trial. The first phase of this trial is a validation phase in which surgeons perform SNB followed by an immediate axillary dissection in a consecutive series of 40 patients with invasive breast cancer. Each surgeon completes a mandatory program of proctored training during this phase. Surgeons who achieve a localization rate of 90% or more and a false-negative rate of 5% or less are eligible to proceed to the randomized phase. All 13 surgeons who completed 40 procedures as part of the validation phase of the ALMANAC trial achieved the set target. This study shows that a standardized training program allows surgeons to achieve a satisfactory localization rate and an acceptable false-negative rate after 40 SNBs.
Key Words: Breast cancer • False-negative rate • Learning curve • Localization rate • Sentinel node biopsy
Sentinel node biopsy (SNB) promises to become the axillary staging procedure of choice for patients with early breast cancer. The fact that the sentinel node can be accurately localized and the concept that the sentinel node reliably predicts the status of the regional lymphatic basin have been confirmed by many studies. However, most authors caution that before SNB is accepted as standard practice in the management of breast cancer it should be validated by randomized trials.1–4 At present many randomized trials are investigating SNB in breast cancer. As with any new surgical technique, SNB is associated with a learning curve. Donald Morton,5 in his first publication describing the technique, emphasized the importance of the learning curve for this new procedure. Since then, there have been various publications on the learning curve in SNB, but there seems to be no consensus as to how many procedures this should involve. The recent consensus meeting in Philadelphia gave figures ranging from 10 to 100 procedures needed to achieve competence.4
The Medical Research Council in the United Kingdom has funded the validation phase of the ALMANAC (Axillary Lymphatic Mapping Against Nodal Axillary Clearance) trial. This is a multicenter, randomized clinical trial involving 14 breast units in the United Kingdom, which include university teaching hospitals and district general hospitals. The trial consists of two phases. Phase 1, a validation phase, has been incorporated into this trial to assess the learning curve associated with this new surgical technique. The second phase of the trial is the randomized phase, comparing SNB with conventional axillary treatment.
PHASE 1: VALIDATION PHASE
All surgeons perform SNB followed by axillary sampling or clearance in 40 patients with invasive breast cancer. Surgeons who achieve a successful localization rate of 90% and a false-negative rate of less than 5% are eligible to proceed to the randomized phase of the trial, which is now in progress.
The false-negative rate is expressed by the falsely negative sentinel node as a percentage of all the patients in a series with a positive axilla. However, there is a problem with this calculation when a predetermined false-negative rate of 5% is set, as there is no way of predicting the percentage of node positivity in each surgeon’s series. If 25% of a surgeon’s 40 cases are node-positive (10 cases), then a surgeon with only 1 false-negative would have a false-negative rate of 10%. However, a surgeon who has 50% node positivity in 40 cases (20 cases) will have the same false-negative rate (10%) with 2 false-negative cases. To remove this bias, the false-negative rate is calculated as a percentage of the total number of patients rather than as a percentage of the positive axillae. Thus, each surgeon is allowed no more than two false negatives in the series of 40 patients. These points were well made in the statement from the Consensus Conference Committee in Philadelphia.4
PHASE 2: RANDOMIZED PHASE
In the randomized phase each patient with an invasive breast cancer is randomized into one of two treatment groups (Fig. 1). The control group undergoes conventional axillary treatment while the study group undergoes SNB. Patients with a positive SNB on paraffin section histology undergo treatment of the axilla with either radiotherapy or a completion axillary clearance, while those patients with a negative SNB have no further treatment to the axilla. Adjuvant treatment in both groups depends on tumor and nodal characteristics.
|
We report on the learning curve of 13 surgeons who completed the validation phase.
METHODS
All surgeons attended a course on SNB, which covered theoretical aspects and featured live surgery demonstrations. Subsequently, all surgeons were proctored in their own institution by the principal investigator of the trial.
Patients with invasive breast cancer, irrespective of the size of the tumor, who were clinically node negative were included in both phases of the study. Pregnant women, patients with clinically involved nodes or multicentric tumors, and those who had undergone previous surgery to the same breast or axilla were excluded.
Following informed consent, each patient had the sentinel node(s) localized by a standardized protocol, with a combination of lymphoscintigraphy, a radiopharmaceutical, and blue dye. The radiopharmaceutical used was Nanocoll (Amersham Health). Two milliliters was injected at four sites around the tumor at a dose of 40 MBq, if injected the day before surgery, or 20 MBq, if injected on the day of surgery. Lymphoscintigraphy was performed approximately 3 hours after injection.
After induction of anesthesia, 2 mL of Patent Blue V dye (distributed by Technical Photo-Systems, UK) diluted to 5 mL was injected peritumorally. A blue-stained lymphatic was followed to a blue-stained node. A hand-held gamma probe was used to confirm the presence of a hot node. If an internal mammary node was seen on lymphoscintigraphy and a corresponding hot spot was found in surgery, an internal mammary node biopsy was performed. After the sentinel node was localized and removed, the patient underwent definitive treatment for the breast cancer, which included axillary sampling or axillary clearance. Axillary sampling is a less invasive procedure that has been validated in large studies; a minimum of four axillary nodes are palpated and removed to stage the axilla. In the United Kingdom, the axilla is staged by axillary sampling or axillary clearance, depending on the surgeon’s preference. The ALMANAC trial has included a representative group of surgeons performing both of these procedures to stage the axilla.
The sentinel node was assessed by standard hematoxylin and eosin staining in accordance with the NHSBSP guidelines. Immunohistochemistry was not used to assess the sentinel node in this study. All data were collected in a standardized case report form and collated centrally by a trial co-coordinator. Hypothesis tests included 
2 (or Fisher exact test when necessitated by small numbers) and Mann-Whitney. Confidence intervals (CIs) of 95% were used for proportions.6
RESULTS
We report on the learning curve of the 13 surgeons who completed the validation phase by July 3, 2001. A further 13 surgeons who completed between 1 and 31 cases (mean, 11.7) by this date are not included. The 520 patients comprised 516 females and 4 males with a mean age of 57.6 years (range, 27 to 82 years). Of the 1095 sentinel nodes identified, 841 (76.8%; CI, 74.2% to 79.2%) were blue and 917 (83.7%; CI, 81.4% to 85.8%) were hot. The combination of the blue dye and radiopharmaceutical gave a successful localization rate of 96.3% (CI, 94.4% to 97.7%) in the whole series of 520 patients. The average time required for SNB was 17.7 minutes (range, 2 to 90 minutes). An average of 2.1 sentinel nodes was identified per patient (range, 1 to 9). The mean tumor size was 19 mm (range, 1 to 100 mm). Of the 520 patients, 169 were node-positive; thus, the axillary positivity rate was 32.5% (CI, 28.6% to 36.6%). There were 10 false negatives, giving a false-negative rate of 5.9% (CI, 3.2% to 10.5%) among positive axillae. Table 1 shows the localization rate and false-negative rate for each of the surgeons in this series. There was little evidence of heterogeneity of these parameters between the 13 surgeons (localization rate: 2 = 10.8, degrees of freedom = 12, P = .54; axillary positivity: 2 = 18.5, P = .10; too few false-negatives for analysis).
|
DISCUSSION
Every new surgical technique is associated with a learning curve, and this is true for SNB in breast cancer. We have learned a great deal about the learning curve in new surgical procedures with the introduction of laparoscopic surgery. Early reports showed a definite learning curve for laparoscopic cholecystectomy; with more experience, surgeons were able to perform this operation safely, quickly, and with a complication rate comparable to that of the traditional open procedure.7 The same is true of laparoscopic gynecological surgery8 and laparoscopic colonic surgery,9 for which studies have shown that with increasing operative experience the results of these laparoscopic procedures are as safe as open procedures and, more important, offer potential additional benefits to the patient.
The importance of the learning curve in SNB was emphasized in Morton’s first publication.5 There are two issues with the SNB learning curve in breast cancer: localization of the sentinel node and pathologic analysis of this node, which can be represented by the false-negative rate (Fig. 2). The literature seems to suggest that with a combination of blue dye and radioisotope the rate of sentinel node localization should be around 90%.10 The false-negative rate, however, has been more variable and difficult to quantify. Guidelines from the American College of Breast Surgeons state that an acceptable false-negative rate would be a skip metastasis rate of 0 to 1 in the first 10 patients with metastasis to the regional basin.11 A further commentary on the false-negative issue recommended calculating the accuracy of SLN biopsy in axillary staging rather than the actual false-negative rate. It has been suggested that the accuracy can be calculated by multiplying the percent risk of nodal metastasis by the false-negative rate.12 In our study there was a widely varying rate of pathologically positive axillae because no upper limit of tumor size was imposed and a variable proportion of screen-detected cancer was included by some surgeons.
|
Our study shows that the learning curve is very short if the surgeon completes a standardized training program. The major element of this program is multidisciplinary proctored training that involves surgeon, nuclear medicine physician, radiologist, and pathology staff. Our results show that this prior teaching has shortened the learning curve in the subsequent audit cases (Fig. 2).
CONCLUSION
The results of the ALMANAC audit program show that when prior training involving a theoretical course and additional proctored training in this new surgical technique are given, there is no statistically demonstrable learning curve for this procedure, apart from the increased failure rate for the surgeon’s first unproctored procedure. We conclude that surgeons should be able to perform this new surgical procedure at a satisfactorily high standard well before completion of 40 cases.
ALMANAC Writing Committee
The committee members are Mr. U. Chetty, Professor P. Ell, Professor L. Fallowfield, Mr. M. Kissin, Professor R. Mansel, Dr. R. Newcombe, and Mr. M. Sibbering.
ALMANAC Trialists Group
The group members are Professor R. Mansel (Principal Investigator), Mr. T. I. Abdullah, Miss E. Anderson, Mr. L. Barr, Professor N. Bundred, Mr. M. Chare, Mr. U. Chetty, Mr. S. Courtney, Mr. D. Crawford, Mr. R. Cummins, Mr. C. Davies, Mr. M. Dixon, Mr. D. England, Mr. J. Geraghty, Mr. H. Holliday, Mr. K. Horgan, Mr. M. Kissin, Mr. M. Lansdown, Mr. T. Mahapatra, Mr. J. McCarthy, Mr. I. Monypenny, Professor M. Perry, Professor J. Robertson, Mr. R. Sainsbury, Mr. M. Sibbering, Mr. D. Sinnett, Miss H. Sweetland, Mr. D. Webster, Mr. J. Winstanley, and Mr. C. Yiangou.
ACKNOWLEDGMENTS
The validation phase of the ALMANAC trial has been funded by the Medical Research Council.
The acknowledgments are available online in the fulltext version at www.annalssurgicaloncology.org. They are not available in the PDF version.
FOOTNOTES
A standardized program of theoretical and surgeon-proctored training in each institution allows surgeons to achieve excellent results for sentinel node biopsy in breast cancer.
Received for publication November 19, 2003. Accepted for publication December 16, 2003.
REFERENCES
This article has been cited by other articles:
 |
|
 |
|
  S. K. Somasundaram, D. W. Chicken, and M. R. S. Keshtgar Detection of the sentinel lymph node in breast cancer Br. Med. Bull., January 3, 2008; (2008) ldm032v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. E. Mansel, L. Fallowfield, M. Kissin, A. Goyal, R. G. Newcombe, J. M. Dixon, C. Yiangou, K. Horgan, N. Bundred, I. Monypenny, et al. Randomized multicenter trial of sentinel node biopsy versus standard axillary treatment in operable breast cancer: the ALMANAC Trial. J Natl Cancer Inst, May 3, 2006; 98(9): 599 - 609. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. B. Roaten, N. Pearlman, R. Gonzalez, R. Gonzalez, and M. D. McCarter Identifying Risk Factors for Complications Following Sentinel Lymph Node Biopsy for Melanoma Arch Surg, January 1, 2005; 140(1): 85 - 89. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
