Application Of Sentinel Node Biopsy In Minimally Invasive Treatment For Early Gastric Cancer

Background and objectiveEarly gastric cancer (EGC) was defined as adenocarcinoma confined to the mucosa or submucosa irrespective of lymph node involvement. It is reported that the incidence of lymph node metastasis determined by histology in mucosal and submucosal gastric cancer is 2% to 4% and 13% to 20%, respectively . It is infered that routine lymphadenectomy may be not necessary for more than 80% early gastric cancer patients. A survey conducted by the Japan Society for Endoscopic Surgery showed that 1428 laparoscopic wedge resections(LWRs). 260 intragastric mucosal resections(IGMRs) were performed between 1991 and 2001 in departments of endoscopic surgery in Japan. The key problem of the operations refered above is that they can't evaluate the status of lymph node metastasis (LNM). We have carried out the application of sentinel node biopsy (SNb) in gastric cancer for nearly 7 years. It is demonstrated that the method was accurate in predicting the status of lymph node metastasis in gastric cancer, especially for early gastric cancer. The objective of this study was to analyze the influencing factors of lymph node metastasis, the result of the application of sentinel node biopsy and the feasibility of this method in early gastric cancer retrospectively. The procedure of laparoscopic-endoscopic rendezvous local resection was carried out in 4 patients with early gastric cancer. The main purpose of this study is to assess the applicability of sentinel node biopsy inminimally invasive treatment for early gastric cancer.Materials and methods1 The clinical characteristics and pathologic features of 257 early gastric cancer patients admitted in our hospital during 1994 and 2006 were analyzed retrospectively. 186 patients were male while 71 patients were female among them. They were (52.85 ±12.59) years old in average. Tumors in 17 patients located in the upper 1/3 of the stomach, 37 patients in the middle 1/3 and 37 patients in the lower 1/3. Tumor sizes were(3.19±1.73)cm in average. All the included patients underwent D2 gastrectomy .2 A retrospective analysis was carried out in 23 early gastric cancer patients who underwent sentinel node biopsy and D2 gastrectomy. Computed tomography (CT) scan was carried out in all the 23 patients, no serosa invasion or distal metastasis was detected. 14 patients were male while 9 patients were female among them. They were (51.39±13.07) years old in average. Tumors in 1 patients located in the upper 1/3 of the stomach. 4 patients in the middle 1/3,18 patients in the lower 1/3. Tumor sizes were (3.22± 1.59) cm in average.The procedure of sentinel node biopsy: 1ml of patent blue vital dye was injected in the peritumoral subserosa after the abdomen was opened. If the tumor could not be palpated, the stomach would be opened or esophago-gastroduodenoscopy would be need to inject the patent blue vital dye in the peritumoral submucosal. All the blue-stain lymph nodes appeared within 5 minutes were taken as sentinel nodes. All the sentinel nodes would be examined by HE. Sentinel nodes that were Immunohistochemistry would be carried out in patients whose sentinel nodes were negative or uncertain.3 Sentinel node rendezvous local resection was carried out in 4 patients with early gastric cancer.Patient selection: Tumor size less or equal to 3cm, protruded or elevated or flat type, definite or skeptical early gastric cancer by preoperative pathology. No serosa invasion or distal metastasis was detected by CT scan. They were all confirmed to be early gastric cancer by frozen section. When a patient is in bad common condition andthe risk for radical gastrectomy is great, the indication would be expanded. All the selected patients were informed the risks of this procedure and they still signed to agree with this method.Procedure: Laparoscopic-endoscopic rendezvous local resection and sentinel lymph node biopsy was carried out in the selected patients, all the specimens and sentinel nodes were examined by frozen section. If they were definited by frozen section and the margins were negative, the gastric incision would be sutured. Then, if all the sentinel nodes were negative, the operation would be completed. If one or more sentinel nodes were positive and the patient's common condition was good, D2 gastrectomy would be carried out. If all the sentinel nodes were shown to be negative by frozen section but one or more of them were proved to be positive be routine pathology, a secondary D2 lymphadenectomy would be proceded.Follow up: All the patients were required to be followed up once 3 months in the recent 3 years, and once 6 months between 2 years and 5 years after operation. The substances paid attention to were symptoms, the abdomen condition, superficial lymph nodes, CEA, CA199 and so on. Esophago-gastroduodenoscopy was required 6 months after operation, and once a year later. 4 Statistics analysisThe relationship between the clinical characteristics and pathologic features and lymph node metastasis was statistically analyzed by SAS system 9.0. Chi-square test and t-test were used in univariate analysis. Logistic regression was used in multivariate analysis. Statistics differences were definited by P < 0.05.Result1 Influencing factors of lymph node metastasis in early gastric cancer 4845 lymph nodes were gained in 257 early gastric cancer patients and 200 lymph nodes of them were positive (4.13%) . Lymph node metastasis were detected in 64 patients( 24.90%), among whom first level (N1) metastasis were detected in 53 patients, first level and second level (N1+N2) metastasis were detected in 7 patients, second level (N2) metastasis alone was detected in 1 patient, all the three level(N1+N2+N3) metastasis were detected in 2 patients and first level plus third level (N1+N3) metastasis was detected in 1 patient. The incidence of lymph node metastasis in mucosal carcinoma was 5.31% (6/113) and it was 40.28% (58/144) in submucosal carcinoma.Lymph node metastasis was seldom detected in mucosal carcinoma whose tumor size was less than 4cm (P<0.05) .The incidence of lymph node metastasis in submucosal carcinoma was obviously higher than mucosal carcinoma (P<0.01) .N2 metastasis was detected only in patients whose tumor size was equal or more than 2cm (P<0.01), N3 metastasis was detected in patients whose tumor size was equal or more than 4cm (P<0.01).The incidence of lymph node metastasis in Undifferentiated early gastric cancer was obviously higher than that in well differentiated type (P<0.05) .The incidence of lymph node metastasis in patients whose tumor located in the lower 1/3 of the stomach was higher than others (P<0.05) .Lymph node metastasis was mainly detected in depressed type. However, there was no obvious difference between depressed type and non-depressed type. The incidence of lymph node metastasis in patients with blood vessel or lymphatic vessel invasion was higher than others, but there was no obvious difference between them.Tumor location, tumor size, depth of invasion, blood vessel invasion or not were independent influencing factors in Multivariate analysis. Tumor size, depth of invasion were the most important influencing factors.2 Retrospective analysis of the application of D2 gastrectomy combined with sentinel node biopsy in early gastric cancer.67 sentinel nodes were detected in the 23 early gastric cancer patients. The detection rate was 100%. Sentinel nodes were positive in 3 patients by routine pathology, and they were the unique positive lymph nodes of all the gained regional lymph nodes. Sentinel nodes in 1 patient was false negative by routine pathology. However, the sentinel nodes of this patient were proved to be positive by immunohistochemistry. Therefore, the sensitivity, specificity, and diagnostic accuracy were all calculated to be 100% in the 23 patients finally.Sentinel nodes of 22 patients located in N1 and 1 patient located in N1+N2. Hence, the incidence of sentinel nodes in N2 was 4.3% (1/23) .The total incidence of lymph node metastasis of the 23 patients was 17.4%(4/23 ).A11 the positive lymph nodes of the 4 patients located in N1.3 Application of sentinel node biopsy in local resection of early gastric cancer. All the 4 patients were proved to be early gastric cancer by frozen section or post-operation routine pathology,all the sentinel nodes and margins were negative, all the sentinel nodes were proved to be negative by immunohistochemistry. Local resection was successfully carried out on the 4 patients,regional lymphadectonay was avoided. All the patients recovered well, no recurrence or metastasis was detected during the follow up period of 6-14 months.Conclusion1 Tumor location, tumor size, depth of invasion, blood vessel invasion or not were independent influencing factors in Multivariate analysis. Tumor size, depth of invasion were the most important influencing factors. The incidence of lymph node metastasis in submucosal carcinoma was obviously higher than mucosal carcinoma. Lymph node metastasis was seldom detected in mucosal carcinoma whose tumor size was less than 4cm, but lymph node metastasis in submucosal carcinoma would be detected in every interval of tumor sizes.2 The sensitivity, specificity, and diagnostic accuracy were all calculated to be 100% in predicting the status of lymph node metastasis. The application of sentinel node biopsy in early gastric cancer is reliable.3 Sentinel node biopy is good at predicting the status of lymph node metastasis. It brings a new method for improving the safety of minimally invasive treatment for early gastric cancer.