OBJECTIVETo compare the anastomotic leakage rates after esophagectomy and reconstruction through different routes for esophageal cancer and analyze the causes for higher anastomotic leakage rate after esophagectomy, systemic lymph node dissection and reconstruction through retrosternal route and its prevention.

METHODSData of 1105 cases of esophagectomy were reviewed retrospectively. Patients in group A (n=229) underwent esophagectomy through left thoracotomy and intrathoracic anastomosis, patients in group B (n=716), esophagectomy through right anterio-lateral thoracotomy and cervical reconstruction through posterior mediastinal route, patients in group C (n=160) esophagectomy, systemic lymph node dissection and cervical anastomosis through the retrosternal route.

RESULTSThe leakage rate was significantly higher (19.4%) in group C than that in group B (11.9%, P< 0.05) and much significantly higher than that in group A (2.2%, P< 0.01). In group C, there was no significant difference in leakage rate between the patients with hand-sewn or mechanical anastomosis (22.2% vs.11.6%, P=0.133), between the patients who had whole stomach or tube-typed gastric reconstruction (25% vs.15.6%, P=0.146). The leakage rate was significantly decreased from 23.3% to 9.1% after prolonged nasal-gastric drainage for seven days (P< 0.05).

CONCLUSIONThe high anastomotic leakage rate after retrosternal reconstruction is mainly due to compression of the stomach in the anterior mediastinum. Prolonged nasogastric drainage is an effective way to decrease the leakage rate after systemic lymphadenectomy.