OBJECTIVETo compare the safety and efficacy between three-field lymphadenectomy and normative Ivor-Lewis two-field lymphadenectomy for thoracic esophageal squamous cell carcinoma METHODS: Clinical data of 375 patients with thoracic esophageal squamous cell carcinoma who underwent three-field lymphadenectomy(3FL) or Ivor-Lewis two-field lymphadenectomy(2FL, Ivor-Lewis) in Fudan University Shanghai Cancer Center during 2013 were retrieved and collected from electronic medical record system. Ninety-one patients received three-field lymphadenectomy (3FL group), including 16 cases of intra-cervical gastro-esophageal anastomosis and 75 cases of intra-thoracic gastro-esophageal anastomosis, while 284 patients received Ivor-Lewis two-field lymphadenectomy (2FL group) with all intra-thoracic gastro-esophageal anastomosis. Short-term outcomes were compared between two groups, including postoperative anastomotic leakage, pneumonia and respiratory failure, chylothorax, reoperation and 90-day death. Total harvested lymph nodes and positive lymph nodes in each group were also compared. A total of 338 patients were enrolled into survival analysis. Survival curve was presented by Kaplan-Meier method.

RESULTSAs compared to 2FL group, the 3FL group had significantly higher ratio of N3 patients [19.8% (18/91) vs. 5.3% (15/284), P=0.000], stageIII( patients [58.2%(53/91) vs. 43.0%(122/284), P=0.007], and upper thoracic cancer patients [12.1%(11/91) vs. 3.5%(10/284), P=0.027]; also the 3FL group had more harvested lymph nodes (40.1±14.6 vs. 25.3±9.4, P=0.000) and more positive lymph nodes (3.3±4.0 vs. 1.7±3.2, P=0.000). With respect to pneumonia and respiratory failure, chylothorax, reoperation and 90-day death, no significant differences were found between the group (P=0.447, P=0.751, P=0.678, P=0.685). The 3FL group had a significantly higher incidence of anastomotic leakage than 2FL group [7.7% (7/91) vs. 1.8% (5/284), P=0.011], while its incidence of intrathoracic anastomosis leakage was 4.0% (3/75), which was not significantly different with 1.8%(5/284) of 2FL group (P=0.372). Median follow-up was 33 months. Overall 1-, 2-, 3-year survival rates were 94%, 81% and 70%, while 1-, 2-, 3-year survival rates of 3FL group were 90%, 73% and 66%, of 2FL group were 95%, 84% and 72%, respectively, without significant differences between the two group(P=0.135). Further subgroup analysis showed that no significant differences of postoperative survival in stage I(, II( and III( patients were observed between the two groups (P=0.541, P=0.511, P=0.402), meanwhile no significant differences of postoperative survival in patients with metastasis and without metastasis were found between the two groups as well (P=0.985, P=0.233).

CONCLUSIONSThree-field lymphadenectomy can be performed with acceptable perioperative morbidity and mortality. The prognosis value of three field lymphadenectomy needs further investigation. Patients with thoracic esophageal squamous cell carcinoma may have favorable survival through normative Ivor-Lewis two-field lymphadenectomy.