Three-field <i>versus</i> two-field lymphadenectomy in transthoracic oesophagectomy for oesophageal squamous cell carcinoma: short-term outcomes of a randomized clinical trial

Abstract

BACKGROUND: The benefit and harm of three-field lymphadenectomy for oesophageal cancer are still unknown. The aim of this study was to compare overall survival and morbidity and mortality between three- and two-field lymphadenectomy in patients with oesophageal squamous cell carcinoma. METHODS: Between March 2013 and November 2016, patients with squamous cell carcinoma of the middle or distal oesophagus were assigned randomly to open oesophagectomy with three-field (cervical-thoracic-abdominal) or two-field (thoracic-abdominal) lymphadenectomy. No chemo(radio) therapy was given before surgery. This paper reports on the secondary outcomes of the study: pathology and surgical complications. RESULTS: Some 400 patients were randomized, 200 in each group. A median of 37 (i.q.r. 30-49) lymph nodes were dissected in the three-field group, compared with 24 (18-30) in the two-field group (P < 0·001). Some 43 of 200 patients (21·5 per cent) in the three-field group had cervical lymph node metastasis. More patients in the three-field group had pN3 disease: 21 of 200 (10·5 per cent) versus 10 of 200 (5·0 per cent) (P = 0·040). The rate and severity of postoperative complications were comparable between the two groups, except that six patients in the three-field arm needed reintubation compared with none in the two-field group (3·0 versus 0 per cent; P = 0·030). The 90-day mortality rate was 0 per cent in the three-field group and 0·5 per cent (1 patient) in the two-field group (P = 1·000). CONCLUSION: Oesophagectomy with three-field lymphadenectomy increased the number of lymph nodes dissected and led to stage migration owing to a 21·5 per cent rate of cervical lymph node metastasis. Postoperative complications were largely comparable between two- and three-field lymphadenectomy. Registration number: NCT01807936 ( https://www.clinicaltrials.gov).