OBJECTIVETo explore the distribution of lymph node metastases, to analyze the cliniopathologic factors of thoracic esophageal carcinoma after curative resection, and to provide the criteria of irradiated region delineation in radiotherapy for esophageal carcinoma.

METHODSThe clinicopathological data of 763 patients who underwent esophagecotomy from Jun 2002 to Jun 2006 were retrospectively analyzed. The regularity of lymph node metastases of thoracic esophageal cancer and clinicopathological factors were stratified and analyzed with SPSS13.0 software.

RESULTSOf the 763 patients, a total of 5846 lymph nodes were dissected with an average of 7.7 lymph nodes in each case. Metastatic lymph nodes were 711, the ratio of metastatic lymph node was 12.2%, and 297 patients had lymph node involved, the lymph node metastasis rate was 38.9%. The metastatic lymph nodes of upper-thoracic esophagus were mainly observed in the supraclavicular and paratracheal regions (P < 0.05), the metastatic lymph nodes of middle-third thoracic esophagus were bidirectional, and those of the lower-third thoracic esophagus mainly metastasized to the regions adjacent to the esophagus, gastric cardia and gastric artery (P < 0.05). Both the metastasis ratio and rate of lymph nodes adjacent to the gastric artery in the lower-thoracic esophageal cancer were significantly higher than those in the middle-third and upper-third thoracic esophageal cancers (P = 0.007, P = 0.001). The multiple factors logistic regression analysis showed that tumor length, depth of tumor invasion, vascular tumor emboli and distant metastasis were major factors for lymphatic metastasis (P < 0.01). For the whole group of patients the lymph node metastatic rate was 28.5% in upper-thoracic esophageal cancer, significantly lower than 38.8% of the lower-thoracic esophageal cancer (P = 0.039) and 43.4% in the middle-thoracic esophageal cancer (P = 0.010). However, the lymph node metastatic rates were 37.0%, 37.9% and 41.4% in the upper-, middle- and lower-thoracic esophageal cancers of the 592 cases receiving left chest notches, with a non-significant difference among them (P = 0.715).

CONCLUSIONThe lesion length, depth of tumor invasion, vascular tumor embolus and distant metastasis are the most important parameters for lymph node metastases. Operative modes have obvious influence on the distribution of regional lymph node metastases. Therefore, in the clinical management, a postoperative prophylactic radiotherapy may be selected according to the tumor length, depth of tumor invasion, vascular tumor embolus and distant lymph node metastasis.