| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
SUPPLEMENT |
From the Department of Surgery, The Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands.
Correspondence: Address correspondence and reprint requests to: Omgo E. Nieweg, MD, PhD, Department of Surgery, The Netherlands Cancer Institute, Plesmanlaan 121, Amsterdam, 1066 CX, The Netherlands; Fax: 31-20-512-2554; E-mail: o.nieweg{at}nki.nl
ABSTRACT
Morton’s original definition of a sentinel node as the first lymph node to receive afferent lymphatic drainage from a primary tumor reflects the concept of stepwise spread of cancer through the lymphatic system. Several new definitions have been developed, based on surgical anatomy and on the technique that is used to find the node. The various definitions of a sentinel node are critically analyzed. Breast cancer surgeons use three different definitions of a false-negative sentinel node biopsy. The best definition appears to be based on the assumption that the procedure is truly positive if either the sentinel node or a suspicious node that is not radioactive or blue contains metastatic disease.
Key Words: Breast neoplasms • Neoplasm staging • Review • Sentinel lymph node
There is a great variability in the number of sentinel nodes that surgeons remove. On a number of occasions, we noticed surgeons taking more than 10 nodes and submitting them as sentinel nodes. At our institution, we remove an average of 2.3 sentinel nodes from melanoma patients. In breast cancer patients, we remove an average of just 2.1 sentinel nodes, despite the fact that we enthusiastically pursue sentinel nodes outside the axilla as well. Apparently, there is controversy over when to designate a lymph node as a "sentinel" node.
Morton and Cochran1 stated that a sentinel node is the initial lymph node to which the primary tumor drains. This definition was slightly modified to identify a sentinel node as a lymph node on the direct drainage pathway from the primary tumor (Fig. 1). This definition requires identification in the nodal basin of a lymphatic channel that originates in the primary tumor area. Unfortunately, this is not always accomplished. Lymphoscintigraphy sometimes shows multiple hot nodes, and early images sometimes fail to show lymphatic channels that can determine the order of drainage. Then, Morton’s definition cannot be applied if no blue channel is found intraoperatively. Thus, the true node at risk may be left behind or too many nodes may be taken. This is a reason why new definitions have been developed (Table 1). 2
|
|
The definition of the sentinel node as the lymph node closest to the primary lesion disregards the physiology of lymph drainage. This is illustrated by the observation that most melanomas of the lower leg and foot have their sentinel node in the groin and not in the popliteal fossa, which is much closer. The node closest to the primary tumor is the first one to be involved only when it receives drainage directly from the primary lesion site.
Late lymphoscintigraphy images may depict more than one node, but they do not visualize the lymphatic channels. Early images do visualize the lymphatic channels and thereby delineate the order of drainage. When multiple nodes are shown, some investigators define the sentinel node as the first lymph node that becomes visible on the lymphoscintigraphic images. The first node that is depicted is directly at risk of receiving tumor cells, but a node that is depicted later on is not necessarily a second-tier node. Multiple lymphatic channels can originate in the same region of the body and run directly to different lymph nodes. These nodes are not always visualized simultaneously. All nodes in direct drainage contact with the primary tumor should be harvested and examined by the pathologist. Therefore, the definition of the sentinel node as the first node that is visualized is too narrow: too few nodes are labeled sentinel nodes, and metastases may be missed.
It is very rewarding to review the lymphoscintigraphic images with the nuclear medicine physician who has performed the study. Interpretation is easy when a single hot spot is shown. Morton’s definition can still be applied if the images show multiple nodes, each with its own afferent channel from the injection site. However, Morton’s definition cannot be applied if multiple nodes are depicted without lymphatic channels. In that situation some people define the sentinel node as the hottest node.3 The hottest node is most often but not always the first to receive tumor cells. Some of the tracer may pass through a sentinel node and move on to subsequent nodes. A large second-echelon node—or one with more active macrophages—may accumulate more radioactivity than a small first-echelon node (Fig. 2). Furthermore, the amount of tracer that is accumulated by a node depends not only on its position in the drainage order but also on the number of lymphatic channels that enter the node and on parameters such as lymph flow rate. One of the reasons for a node to receive a sparse lymph supply is obstruction of flow by metastatic disease (Fig. 3). This phenomenon is illustrated by recent studies showing that the involved node was the most radioactive node in not more than 80% of patients in whom more than one sentinel node was identified.4,5 A node may contain so much tumor that it does not take up any of the tracer. Fortunately, nodes that contain large amounts of metastatic disease are likely to be identifiable on ultrasound or may be palpable intraoperatively.6
|
|
Some surgeons rely on their gamma ray detection probe and pursue (all) radioactive nodes. The disadvantage of the probe is that it cannot visualize the lymphatic channel, which means that it cannot distinguish first-tier nodes from second-tier nodes. Secondary nodes may then be removed unnecessarily. Also, 15% to 30% of the sentinel nodes in breast cancer patients are only blue and not radioactive; these would be missed.7,8
A more sophisticated approach is to define a sentinel node by its radioactivity relative to the background—the sentinel node-to-background ratio. A problem with this definition is that the amount of radioactivity that travels to a lymph node varies with the type of colloid particles that are used, their size and stability, their surface characteristics, the size of the lymph node, and the lymph flow rate.9,10 The background count rate is also variable and depends, for instance, on the proximity of the injection site, where most of the radioactivity still lingers.
Some surgeons rely on the blue dye and remove every blue-stained node. However, dye is not retained in a node. It flows through and travels to the next node in line. So, at a certain point in time, there will be a string of blue nodes, of which only the first one is directly at risk of harboring tumor cells.
DISCUSSION
The sentinel node is not necessarily the nearest node, nor is it only the one depicted first on the images. Not every radioactive node is a sentinel node, and not every sentinel node is radioactive. Not every blue node is a sentinel node, and not every sentinel node is blue. Morton’s original definition of a sentinel node as a node that receives afferent lymphatic drainage directly from a primary tumor best reflects the concept of stepwise spread of cancer through the lymphatic system. However, this definition is based on a concept and it is not always helpful when the nuclear medicine physician and the surgeon are confronted with a clinical situation that is not as clear-cut as theory would suggest. Surgeons cling to seemingly watertight guidelines. They particularly like the definition based on the sentinel-node-to-background ratio. It tells them exactly what to do: measure the number of counts in the sentinel node, measure the number of counts in the surrounding tissue, and then divide the two. If the outcome is greater than a certain reference value, then the node is a sentinel node. If the outcome is less than that particular reference value, then the node is not a sentinel node. Such a definition is hard to resist because it gives the surgeon clear instructions and there is no doubt about the outcome.
We do not mean to say that this last definition and the others are useless. They are correct most of the time, of course. But what are they? One is an anatomical definition that does not take into consideration the physiology of lymph drainage. The others simply reflect aspects of the technology applied to find the node on a direct drainage pathway.11 One should not narrow-mindedly follow these definitions and thereby lose sight of the concept of solid neoplasms spreading in a stepwise fashion through the lymphatic system. Occasionally, the shortcomings of the alternative definitions lead to designation of a node as a sentinel node, although it is not directly at risk of harboring tumor cells.
The best approach is to keep Morton’s definition in mind as the definition that reflects the concept and to use all the available detection techniques in the repertoire to find that node. The scintigraphic images indicate the area to explore and oftentimes the order in which the nodes receive drainage. Careful dissection of the blue lymphatic channels lays out the drainage pattern and identifies the node(s) that receive(s) drainage directly from the primary lesion. If the early scintigraphic images are inconclusive and the blue dye approach fails, it is best to err on the safe side: take the probe and remove the radioactive nodes that potentially receive direct drainage from the primary lesion site.
A number of surgeons are concerned about whether the correct node is generally identified as a sentinel node.2,3,11–14 Pathologists also wonder whether the true sentinel node has been removed. They traditionally feel it is their duty to keep surgeons in check. Pathologists do not look for radioactivity, and the blue dye has been flushed from the node by the time it reaches their microscope. How, then, can they be certain that the correct node was taken? A simple approach is being pioneered by Cochran and involves addition of charcoal particles to the tracers.15,16 These particles remain in the lymph node. They are identifiable with the microscope and allow the pathologist to determine whether the correct node was taken. Although this approach will provide both the pathologist and the surgeon with valuable information, it still leaves the surgeon with the difficult decision of which node(s) to take.
Definitions of a False-Negative Sentinel Node Biopsy Specimen
What is a false-negative sentinel node biopsy specimen from a breast cancer patient, and how should the rate of false negativity be calculated? A mistake that is sometimes made is to calculate the rate of false-negative results over the entire group of patients, both those who are axilla-positive and those who are axilla-negative. Because it is not possible to miss metastasis in a patient who has no metastasis, the false-negative rate should be calculated only over the group of axilla-positive patients.
The choice of definition has substantial implications for the false-negative rate. Different investigators have different opinions on when the result of a sentinel node procedure is falsely negative. Issues of debate regarding a patient with a tumor-free sentinel node include the following questions. What if a palpable, nonradioactive, unstained lymph node found during operation is tumor-positive? What if additional sections of the sentinel node subsequently show metastatic disease? What if the pathologist finds a tumor-positive nonsentinel node in the axillary tail of a simple mastectomy specimen? As illustrated in Fig. 3, the sentinel node can sometimes be invaded by tumor, blocking the lymph supply. If the tracers cannot reach this node, they will drain to a "neosentinel" node. Should the procedure be considered falsely negative if this neosentinel node is tumor-free?
In our opinion, there can be three credible definitions of a false-negative finding in sentinel node procedures. The first one follows the original definition of the sentinel node as described above. This node is identified by lymphoscintigraphy, a probe, or blue dye. The identification of this particular lymph node is the sole purpose of this procedure. The result of this procedure would be falsely negative if this sentinel node is disease-free at initial pathology evaluation but a tumor is identified in any axillary lymph node at any time.
The second definition is based on the assumption that the procedure is truly positive if, in addition to the tumor-free sentinel node found during the standard procedure, an additional node is identified that proves to be tumor-positive. For instance, identification of a palpable, tumor-positive, nonsentinel node in combination with a tumor-free sentinel node would not be a false-negative finding if palpation of the biopsy wound is part of the protocol. Coincidental removal of other nodes is obviously not part of such a protocol. Therefore, the finding of other nodes can result in a false-negative outcome. Also, a pathology sampling error would result in a false-negative procedure, because the pathologist must perform additional serial sectioning to find tumor cells in the sentinel node.
One can also call the results of a sentinel node procedure falsely negative only if an axillary recurrence develops during follow-up. In this case, the purpose of the sentinel node procedure is to detect metastatic disease in the axilla. This procedure involves excision of the sentinel node and removal of other nodes that look suspicious. Unintended excision of tumor-positive nodes in the axillary tail of a total mastectomy specimen or together with the sentinel node will not cause the result to be falsely negative.
According to the first definition, we would have eight false-negative cases in the 600 sentinel node biopsy procedures performed at our institution. A suspicious palpable but unstained and nonradioactive node was tumor-positive in four of these procedures, whereas the sentinel node was tumor-free. Palpation of the biopsy wound at the time of surgery is part of our standard procedure, so if we were to use the second definition, only four cases would have a false-negative result. Only two patients developed an axillary recurrence, after 11 and 22 months, respectively, so the third definition of a false-negative finding would improve our results dramatically.
CONCLUSIONS
The first definition of a false-negative result as discussed above has been universally used in learning phase studies. Now that lymphatic mapping is the sole method used for lymphatic staging, we have seen a shift toward the use of the third definition. In our opinion, this definition is too broad. The second definition encourages removal of apparently tumor-positive nonsentinel nodes and discourages the indiscriminate removal of additional nonsentinel nodes, which is tempting when following the last definition. Therefore, the second definition best serves the clinical purpose, which is to identify metastatic disease in the least disruptive manner.
FOOTNOTES
Breast cancer surgeons use three different definitions of a false-negative sentinel node. The best definition appears to be based on the assumption that the procedure is truly positive if either the sentinel node or a suspicious node that is not radioactive or blue contains metastatic disease.
Received for publication November 14, 2003. Accepted for publication December 9, 2003.
REFERENCES
This article has been cited by other articles:
 |
|
 |
|
  I. M. C. van der Ploeg, B. B. R. Kroon, N. Antonini, R. A. V. Olmos, E. J. T. Rutgers, and O. E. Nieweg Axillary and Extra-axillary Lymph Node Recurrences after a Tumor-Negative Sentinel Node Biopsy for Breast Cancer Using Intralesional Tracer Administration Ann. Surg. Oncol., April 1, 2008; 15(4): 1025 - 1031. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. P. Cronin-Fenton, L. A. Ries, L. X. Clegg, and B. K. Edwards Rising Incidence Rates of Breast Carcinoma With Micrometastatic Lymph Node Involvement J Natl Cancer Inst, July 4, 2007; 99(13): 1044 - 1049. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Wada, S. Imoto, C. Yamauchi, T. Hasebe, A. Ochiai, and S. Ebihara Correlation Between Concordance of Tracers, Order of Harvest, and Presence of Metastases in Sentinel Lymph Nodes With Breast Cancer Ann. Surg. Oncol., June 1, 2005; 12(6): 497 - 503. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S H Kim, J Shim, C K Kim, J Machac, and B R Krynyckyi Reverse echelon node and a lymphatic ectasia in the same patient during breast lymphoscintigraphy: the importance of injection and imaging technique Br. J. Radiol., December 1, 2004; 77(924): 1053 - 1056. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
