[Survival comparison of Siewert II adenocarcinoma of esophagogastric junction between transthoracic and transabdominal approaches:a joint data analysis of thoracic and gastrointestinal surgery].

OBJECTIVE: To compare the long-term survival outcomes of Siewert II adenocarcinoma of esophagogastric junction (AEG) between transthoracic (TT) approach and transabdominal (TA) approach.

METHODS: The databases of Gastrointestinal Surgery Department and Thoracic Surgery Department in West China Hospital of Sichuan University from 2006 to 2014 were integrated. Patients of Siewert II AEG who underwent resection were retrospectively collected.

INCLUSION CRITERIA: (1) adenocarcinoma confirmed by gastroscopy and biopsy; (2) tumor involvement in the esophagogastric junction line; (3) tumor locating from lower 5 cm to upper 5 cm of the esophagogastric junction line, and tumor center locating from upper 1 cm to lower 2 cm of esophagogastric junction line; (4)resection performed at thoracic surgery department or gastrointestinal surgery department; (5) complete follow-up data. Patients at thoracic surgery department received trans-left thoracic, trans-right thoracic, or transabdominothoracic approach; underwent lower esophagus resection plus proximal subtotal gastrectomy; selected two-field or three-field lymph node dissection; underwent digestive tract reconstruction with esophagus-remnant stomach or esophagus-tubular remnant stomach anastomosis above or below aortic arch using hand-sewn or stapler instrument to perform anastomosis. Patients at gastrointestinal surgery department received transabdominal(transhiatal approach), or transabdominothoracic approach; underwent total gastrectomy or proximal subtotal gastrectomy; selected D1, D2 or D2 lymph node dissection; underwent digestive tract reconstruction with esophagus-single tube jejunum or esophagus-jejunal pouch Roux-en-Y anastomosis, or esophagus-remnant stomach or esophagus-tubular remnant stomach anastomosis; completed all the anastomoses with stapler instruments. The follow-up ended in January 2018. The TNM stage system of the 8th edition UICC was used for esophageal cancer staging; survival table method was applied to calculate 3-year overall survival rate and 95% cofidence interval(CI); log-rank test was used to perform survival analysis; Cox regression was applied to analyze risk factors and calculate hazard ratio (HR) and 95%CI.

RESULTS: A total of 443 cases of Siewert II AEG were enrolled, including 89 cases in TT group (with 3 cases of transabdominothoracic approach) and 354 cases in TA group. Median follow-up time was 50.0 months (quartiles:26.4-70.2). The baseline data in TT and TA groups were comparable, except the length of esophageal invasion [for length <3 cm, TA group had 354 cases(100%), TT group had 44 cases (49.4%), χ²=199.23,P<0.001]. The number of harvested lymph node in thoracic surgery department and gastrointestinal surgery department were 12.0(quartiles:9.0-17.0) and 24.0(quartiles:18.0-32.5) respectively with significant difference (Z=11.29,P<0.001). The 3-year overall survival rate of TA and TT groups was 69.2%(95%CI:64.1%-73.7%) and 55.8% (95%CI:44.8%-65.4%) respectively, which was not significantly different by log-rank test (P=0.059). However, the stage III subgroup analysis showed that the survival of TA group was better [the 3-year overall survival in TA group and TT group was 78.1%(95%CI:70.5-84.0) and 46.3%(95%CI:31.0-60.3) resepectively(P=0.001)]. Multivariate Cox regression analysis revealed that the TT group had poor survival outcome (HR=2.45,95%CI:1.30-4.64, P=0.006).

CONCLUSION: The overall survival outcomes in the TA group are better, especially in stage III patients, which may be associated with the higher metastatic rate of abdominal lymph node and the more complete lymphadenectomy via TA approach.