| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
SUPPLEMENT |
From the Department of Surgery (MB, HJS, AS, MF, JRS), Department of Nuclear Medicine (MP), and Institute of Pathology (JN), Klinikum rechts der Isar, Technical University of Munich, Munich, Germany.
Correspondence: Address correspondence and reprint requests to: Maria Burian, MD, Chirurgische Klinik und Poliklinik, Klinikum rechts der Isar, Ismaninger Str. 22, 81675 Munich, Germany; Fax: 49-89-4140-4940; E-mail: burian{at}nt1.chir.med.tu-muenchen.de
ABSTRACT
Because of surveillance strategies in patients with known Barrett’s esophagus, more patients with high-grade dysplasia or early cancer in the distal esophagus and at the esophagogastric junction are identified. The need for and extent of lymphadenectomy in such patients are controversial. The technique of sentinel lymph node dissection (SLND) to diagnose early lymphatic spread is applied increasingly in tumors of the gastrointestinal tract. The poorly defined lymphatic drainage of the esophagogastric junction has so far prevented many investigators from performing SLND in tumors of this anatomic region. We report the first results of SLND in Barrett’s and cardia cancer. The preliminary experience indicates that the method is, even in this anatomical area, feasible and yields good results in early tumors. In advanced tumors, the method lacks sensitivity. Mapping should be done with blue dye and a radiocolloid. The concept of sentinel lymph node mapping and detection thus may open the door to individualized therapy for patients with high-grade dysplasia in a Barrett’s esophagus or with early Barrett’s and cardia cancer.
Key Words: Barrett’s carcinoma • Cardia cancer • Sentinel lymph node dissection • Lymphadenectomy
A striking epidemiological trend in the Western world is the rise in incidence of adenocarcinoma of the esophagus (Barrett’s carcinoma) in white males. Because Barrett’s carcinoma has the most rapidly increasing incidence among gastrointestinal tumors, this tumor entity is of major scientific and clinical interest. Numerous research programs on oncogenesis and prediction of malignant degeneration are under way. Furthermore, the identification of a precursor lesion like Barrett’s esophagus and the accessibility of the esophagus for endoscopy and biopsy allow development of specific surveillance programs and early detection. Although the overall prognosis of adenocarcinoma of the esophagus is poor, surgery can provide cure of early-stage tumors.1 In gastric cancer, which is decreasing in incidence worldwide, we see in Western countries a shift in the location of the tumors. While the incidence of tumors of the gastric antrum is decreasing, there is a marked increase in tumors of the cardia and the proximal stomach.
With this background in mind, sentinel lymph node dissection (SLND) in Barrett’s and cardia cancer is a very attractive approach. Because of surveillance programs, we are treating more and more patients with high-grade dysplasia (high-grade intraepithelial neoplasia) and early-stage (uT1A+B) tumors. Japanese studies in gastric cancer2,3 and American studies in colon cancer4,5 indicate that the best results and most convincing impact of SLND therapy can be expected in early-stage tumors. Lymph node status is the major prognostic factor after complete resection of these tumors. Lymphatic spread is closely correlated to the pT category of the primary tumor, starts only after infiltration of the submucosa, and is initially limited to regional lymph nodes.6 Thus, the technique of SLND might set the stage for a tailored resection and lymphadenectomy strategy in Barrett’s and cardia cancer.
DEVELOPING A TAILORED APPROACH TO BARRETT’S CARCINOMA
The traditional radical surgical approaches to adenocarcinoma of the distal esophagus or cardia include total esophagogastrectomy, subtotal esophagectomy with proximal gastric resection, or total gastrectomy with distal esophageal resection, all in combination with an extended lymphadenectomy.6,7 The marked increase of early tumors located in the distal esophagus or at the esophagogastric junction has recently prompted surgeons and interventional endoscopists to search for less-invasive approaches.8 The poor quality of life after extended esophageal and gastric resections9 and the morbidity of these procedures, which still exceeds 40% even in experienced hands,6–8 clearly indicate the need for a less-invasive but curative approach to early-stage adenocarcinoma of the distal esophagus or gastric cardia.8
On the basis of these concepts, we have changed our approach for early adenocarcinoma of the distal esophagus and cardia to a limited (distal) esophagectomy and proximal gastrectomy with a regional lymphadenectomy of the lower mediastinum and around the celiac axis. In order to avoid postoperative reflux, reconstruction is done with a jejunal interposition (Fig. 1).8,10
|
pT1b category). Topographic distribution of positive lymph nodes followed certain rules; skip metastases were rare.11 Thus, a tailored lymphadenectomy based on the sentinel node concept appeared to be possible.
|
Twenty patients with adenocarcinomas of the esophagogastric junction (AEG types I, II, and III) received a preoperative injection of technetium colloid. Ten of the patients also received an intraoperative injection of blue dye. There were 12 patients with pT1 tumors, 7 patients with pT2 tumors, and 1 patient with a pT3 tumor. Ten patients had AEG I tumors (Barrett’s cancer), six patients had AEG II tumors (cardia cancer), and four patients had AEG III tumors (subcardial tumor).
Sentinel node identification was possible in 17 of 20 patients (85%). Blue dye marking was easily possible for AEG III tumors because most of the lymphatic flow could be seen after laparotomy and opening of the lesser sac. In patients with Barrett’s carcinoma and true cardia cancer, sentinel nodes in the lower mediastinum were not visible with the dye and could be detected only with radioactive colloid, even after wide opening of the esophageal hiatus. This might be because most patients with these tumors were markedly obese.
Sentinel node identification was possible in 90% of T1 tumors (9 of 10) and 80% of T2/3 tumors (8 of 10). In advanced tumor stages the method failed to indicate metastatic disease. This might be due to macroscopically involved nodes or an extensive lymphangiosis carcinomatosa, which is often present in advanced-stage AEG I and II tumors. Metastases were found at the right and left cardia site as well as at the lesser curvature in Barrett’s cancer.
Lymph node status was correctly predicted in 100% of AEG I tumors (9 of 9) and 75% of AEG II and III tumors (6 of 8). Overall, sentinel node detection was feasible in 85% of the patients. In our preliminary experience the grading of the primary tumor had no effect on success of sentinel node detection and prediction of the lymph node status.
CONCLUSION
Sentinel lymph node mapping is feasible in Barrett’s and cardia cancer, despite the anatomic complexity of this area of the gastrointestinal tract. The overall detection rate of only 85% might be due to our learning curve and/or technical problems caused by the obesity of the patients. As in other gastrointestinal tumors, especially gastric cancer, lymphatic drainage does not follow distinct anatomical rules. In Barrett’s cancer, drainage is easier to follow after radioactive labeling of the primary tumor. Therefore, the combination of blue dye and radiocolloid injection should be used to map lymphatic drainage in this anatomical area.12 As in other tumor entities, the SLND detection rate in Barrett’s and cardia cancer is much better when tumors are T1 and T2 categories. In advanced tumors, in which invasion of the lymphatic vessels (also termed lymphangiosis) is often present, we had a 25% rate of false-negative results. In these cases the method failed to identify histologically involved lymph nodes.
Regarding sentinel node detection in Barrett’s cancer, there are two other problems to overcome. The area of malignant degeneration is sometimes difficult to see by endoscopy. HGD and T1 carcinomas usually have multicentric locations, which make preoperative and intraoperative mapping more difficult. However, SLND is important not only to determine the exact lymph node status but also to undertake a more precise lymphadenectomy with all marked lymph nodes. Lymphadenectomy can be avoided in node-negative cases, and standard lymphadenectomy can be improved to include nodes in atypical positions.
With the establishment of the sentinel node technique for adenocarcinoma of the esophagogastric junction, individualized therapeutic concepts can be developed. If laparoscopic sentinel node detection is performed and if the door is opened for minimally invasive resection or laparoscopy-assisted limited resection, a tailored concept for early Barrett’s esophagus may become possible.13,14 Figure 2 shows an algorithm to be considered for individualized therapy of HGD in the distal esophagus and early Barrett’s cancer.
|
ACKNOWLEDGMENTS
Supported by Deutsche Krebshilfe (grant number 70–2789 Si 3).
The acknowledgments are available online in the fulltext version at www.annalssurgicaloncology.org. They are not available in the PDF version.
FOOTNOTES
The concept of sentinel lymph node mapping and detection may open the door to individualized therapy for patients with high-grade dysplasia in Barrett’s esophagus or with early Barrett’s and cardia cancer.
Received for publication November 20, 2003. Accepted for publication December 10, 2003.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
