| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Division of Cardiothoracic Surgery, Washington University School of Medicine, St. Louis, Missouri
b Mayo Clinic College of Medicine, Rochester, Minnesota
c Department of Thoracic Surgery, Vanderbilt University Medical Center, Nashville, Tennessee
Accepted for publication April 7, 2008.
* Address correspondence to Dr Meyers, Division of Cardiothoracic Surgery, Washington University School of Medicine, Suite 3108 Queeny Tower, 1 Barnes-Jewish Hospital Plaza, St. Louis, MO 63110 (Email: meyersb{at}wustl.edu).
Presented at the Forty-fourth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 28–30, 2008.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Methods: The operative and pathology reports of patients who underwent staging by computed tomography and fluorodeoxyglucose-positron emission tomography and subsequent resection for esophageal cancer (multiinstitutional American College of Surgeons Oncology Group Z0060 trial) were analyzed.
Results: One hundred forty-five patients underwent resection. Operative and pathology reports were unavailable in 11 patients. The results of the remaining 134 resections (Ivor-Lewis, n = 64; transhiatal, n = 59; other, n = 11) were reviewed. Overall, 13 ± 9 (mean ± standard deviation) lymph nodes were evaluated per patient. More lymph nodes were evaluated in patients undergoing Ivor-Lewis (15 ± 9) than transhiatal esophagectomy (9 ± 7; p < 0.001). The mean number of distinct lymph node stations analyzed per patient was 3 ± 2. In 38% (51 of 134) of patients the nodes attached to the specimen were evaluated without any distinction among nodal stations. The practice of submitting named packets of nodal material resulted in 16 ± 9 nodes per case, as opposed to the practice of submitting an entire specimen for the pathologists to dissect, which yielded 10 ± 8 nodes (p < 0.001).
Conclusions: There is considerable variability and room to improve in the extent of resection and pathologic evaluation of esophagectomy specimens. A uniform standard for esophageal cancer resection is warranted to improve the precision and value of pathologic staging.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
In an effort to refine the prognostic ability of surgical treatment and to improve the likelihood of surgical cure, some investigators have advocated changes to the TNM classification system. Given the worse outcome associated with nodal involvement, Rice and colleagues [1] have recommended subclassifying the N-descriptor to describe 1 or 2 positive nodes as N1 and 3 or more positive nodes as N2. Other investigators have proposed that not only is the number of involved lymph nodes important, but the extent of lymphadenectomy is important as well. Rizk and colleagues [2] have proposed that at least 18 lymph nodes be analyzed to define adequate lymphadenectomy. There are currently no American Joint Committee on Cancer standards as to the extent of surgical lymphadenectomy.
For any proposed revision to the present TNM classification system to be of benefit for the evaluation and treatment of esophageal cancer, the extent of resection must be uniform, and the pathologic analysis of surgical specimen must be precise. Any proposed standard must be widely achievable. In this study, we sought to describe the current practice of esophageal resection in the United States, by reviewing the pathologic results and surgical practices of institutions participating in the American College of Surgeons Oncology Group Z0060 trial [3].
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
The investigators in this report were blinded to patient demographics and identifiers. However, imaging reports, operative records, and the pathologic reports were available for review.
Surgical resection and pathologic evaluation was assessed for type of resection and extent of lymphadenectomy. Nodal stations were defined by the American Joint Committee on Cancer's Cancer Staging Manual, sixth edition [4]. Credit was given to the surgeon for lymph node identification if the surgeon identified and separately submitted two or more lymph nodes or nodal stations for analysis. Otherwise, the pathologist was given credit for nodal identification. If no data were available for location of nodes obtained from a resected specimen, we assumed that level 8 (paraesophageal) and level 17 (left gastric nodes) were analyzed.
The two-sample rank sum test was used to compare groups of subjects. In all cases, probability values of less than 0.05 were considered statistically significant.
This study was approved by the Washington University School of Medicine institutional review board.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Of the patients analyzed, 64 underwent Ivor-Lewis resection, 59 underwent transhiatal esophagectomy, 9 patients underwent a McKeown type operation with laparotomy, thoracotomy, and neck incision, 1 patient underwent esophagectomy by left thoracoabdominal incision, and 1 patient underwent total gastrectomy.
The tumors were located in the middle esophagus in 9 patients, the lower esophagus in 64 patients, and the gastroesophageal junction in 61 patients. The final histologic type was adenocarcinoma in 115 patients, squamous cell carcinoma in 18 patients, and gastric adenocarcinoma in 1 patient. Forty-two patients received some form of neoadjuvant chemotherapy or chemoradiotherapy, whereas 92 did not.
Overall, 13 ± 9 (median = 11) lymph nodes were evaluated per patient. In 11% (15 of 134) of patients, three or fewer lymph nodes were analyzed, and in 4% (5 of 134), no lymph nodes were analyzed. In 36% (48 of 134), 15 or more lymph nodes were analyzed (Fig 1). The pathologist was credited with lymph node identification in 18 of the patients having 15 or more lymph nodes analyzed. Neoadjuvant therapy did not affect the number of lymph nodes evaluated (neoadjuvant, 13 ± 10; median = 9.5 versus no therapy, 13 ± 8; median = 12; p = 0.508). More lymph nodes were evaluated in patients undergoing Ivor-Lewis (15 ± 9; median = 13) than transhiatal esophagectomy (9 ± 7; median = 9; p < 0.001).
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Although a few surgeons routinely identify multiple lymph node stations from the resected specimen and submit them for evaluation, in 38% of specimens there was no distinction of lymph node stations. Surprisingly, 4% of patients had no lymph nodes analyzed, and 11% had three or fewer lymph nodes analyzed. Only 36% had the minimal 15 recommended lymph nodes analyzed using the American Joint Committee on Cancer gastric cancer evaluation guidelines [4]. In 38% of the specimens having at least 15 lymph nodes analyzed, the pathologist was credited with lymph node identification.
Given the variability in surgeon and pathologist practices, staging using the current TNM classification may not be reliable. Not only are a limited number of lymph nodes being analyzed, we report that only 22% of patients had celiac nodes identified and evaluated. Most of the tumors in our series were at the gastroesophageal junction or lower esophagus. For tumors of the lower esophagus and gastroesophageal junction, celiac nodes define M1a disease [1]. Without the surgeon's participation in labeling these nodes, the pathologist is unable to report an accurate M1 descriptor because only the surgeon is able to segregate a left gastric node from a celiac node. This distinction may mean the difference between stage IIB esophageal cancer and stage IV esophageal cancer using current American Joint Committee on Cancer TNM descriptors.
Although we demonstrate a difference in the number of lymph nodes analyzed in Ivor-Lewis resection specimens compared with transhiatal esophagectomy specimens, we cannot conclude the superiority of one approach to another. It is possible to evaluate more than 15 nodes per specimen with either approach. Feith and colleagues [5] demonstrated no difference in survival with either operative approach in patients with gastroesophageal junction tumors. In their series, all patients underwent an extended lymphadenectomy in the abdominal compartment and lower thoracic compartment. An R0 resection is, however, essential to improved outcome [5]. At present, there are no guidelines as to what constitutes an R0 resection in the setting of N1 disease. A number of investigators have favored the concept of a metastatic lymph node ratio. Tachibana and colleagues [6] suggest a metastatic node to normal node ratio of less than 0.09 to predict a more favorable outcome, whereas Rizk and coworkers [2] suggest fewer than 4 metastatic nodes to predict a more favorable outcome when at least 18 lymph nodes are analyzed.
Our study has a number of limitations. Data were collected from operative notes and pathology reports in a retrospective manner. Both reports may be affected by underreporting of the extent of dissection and analysis. A specimen may not have had all the available lymph nodes analyzed if the initial nodes examined were positive for malignancy. Similarly, a surgeon may be more or less likely to perform an extensive lymphadenectomy or lymph node identification based on the clinical stage of the tumor. At present there are no guidelines as to what constitutes adequate lymphadenectomy in patients with esophageal cancer.
We conclude from our analysis that for optimal staging of a patient with cancer of the esophagus, both the surgeon and pathologist must be committed to accurate and detailed analysis of the surgical specimen. Although there are no specific guidelines as to the number of lymph nodes to identify, a more extensive analysis leads to improved prognostic ability, even when some of the lymph nodes are involved with cancer. Multiple investigators have reported the prognostic value of the ratio of positive lymph nodes to the overall number of lymph nodes resected [2, 5, 6].
The best results occur when the surgeon focuses attention to lymph node identification. The simple practice of the surgeon identifying individual nodal stations and submitting lymph node packets separately improves the number of lymph nodes analyzed and the accuracy of the TNM classification.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
DR VEERAMACHANENI: In this trial I don't have data on the volume done by a specific institution. Doctor Putnam, you can correct me if I'm wrong, in order to participate, the surgeon had to be a member of the ACOSOG (American College of Surgeons Oncology Group) study group and therefore had some degree of expertise in the area.
DR ROCCO: One more thing. Thank you for providing the manuscript in advance. I noticed that one in five of the resected specimens contained the celiac nodes. Would you comment on that? Obviously, there could be a prognostic implication.
DR VEERAMACHANENI: Most of the tumors in our series were of the lower esophagus and the GE (gastroesophageal) junction. Comparing similar large series of resected specimens, and specifically Dr Siewert and colleagues' published experience, the predominance of positive lymph nodes are going to occur in the abdomen. Doctor Siewert's group advocates an extensive D2 type of lymphadenectomy in the abdominal compartment. They report a low incidence of nodal involvement in the upper chest. Based on such data, one could argue that a more extensive abdominal compartment dissection would have some prognostic benefit. The current TNM descriptor differentiates whether an involved node is a celiac node versus another involved node. The extent of dissection and the impact of other nodal stations are not defined. This is a point of controversy.
DR KAMAL G. KHALIL (Houston, TX): This is highly controversial. I have a disclaimer. I don't do transhiatal approach. I notice in your description that, on average, transhiatal had 9 nodes or less and Ivor-Lewis had 15 or more. A, does that have any implication about the use of lymphadenectomy in general in terms of prognosis and cure, and, B, if you do transhiatal, how could you define the location of the nodes within the chest? Is that technically possible?
DR VEERAMACHANENI: Thank you very much for your comments.
Regarding the question of whether there is an advantage of one set of incisions to another, my understanding of the literature is that one cannot make a very strong argument for one set of incisions over another. What seems to be most important is achieving an R0 type resection. By R0, I do not mean just surgically negative margins, but a wide enough lymphadenectomy so that even if one or two lymph nodes are involved, we can conclude that the level of nodal involvement is very small and therefore we have not left behind microscopic disease.
In terms of defining the location, I think how we do the resection is going to depend largely on where the tumor is. For a gastroesophageal junction tumor, the incidence of high intrathoracic nodal involvement in the largest published series (Siewert) is rather low. This is a controversial area since proponents of three-field dissection report as high as 25% to 30% nodal involvement in the neck when there is a GE junction tumor.
My own personal bias would be to do an Ivor-Lewis resection on everyone, and this is my own personal bias. I would strive to identify the nodal stations and submit as much as possible for lymph node analysis.
DR JOE B. PUTNAM (Nashville, TN): This paper reflects the variability in practice that exists even within dedicated surgeons performing esophagectomy, and highlights the need for a systematic lymph node dissection for esophageal carcinoma just as we have for lung cancer. Obviously the case volumes for esophageal cancer are much less and we have fewer operations being done for esophageal cancer as are done for lung cancer. However, this presentation identifies the critical role of the surgeon in identification and labeling of specific lymph nodes during the dissection. The presentation further highlights the need for the surgeon to identify these lymph nodes at the time of dissection and to separate them for pathology review.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Deeter, R. Dorer, M. K. Kuppusamy, R. P. Koehler, and D. E. Low Assessment of Criteria and Clinical Significance of Circumferential Resection Margins in Esophageal Cancer Arch Surg, July 1, 2009; 144(7): 618 - 624. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
