BACKGROUND: Although patients with esophageal cancer (EC) often develop lymph node metastases in the cervical and recurrent laryngeal (CRL) distribution, lymphadenectomy in this field is rarely performed. The purpose of this study was to determine factors associated with CRL node positivity and to determine the appropriate indications to perform a "three field" lymphadenectomy. METHODS: In a retrospective review, EC patients who underwent three-field lymphadenectomy were analyzed. Predictors of positive CRL nodes were examined univariately, then selected for inclusion in a multivariate logistic regression model. RESULTS: From 1994 to 2009, 185 patients had a three-field lymphadenectomy, of whom 46 patients (24.9%) had positive CRL nodes. Final pathology stages (seventh edition) were I in 24 patients, II in 43, III in 109, and IV in 1 patient. Eight patients had a major pathologic response after induction therapy. On univariate analysis, variables significantly associated with positive CRL nodes included squamous cell histology, proximal location, advanced clinical presentation, the presence of clinical nodal disease, higher pT classification, and higher pN classification. There was no reduction in the rate of positive CRL nodes after induction chemotherapy. On multivariate analysis, higher pN classification (adjusted odds ratio 16.25, 95% confidence interval: 5.40 to 48.87; p < 0.0001) and squamous histology (adjusted odds ratio 6.04, 95% confidence interval: 2.21 to 16.56; p < 0.0001) predicted positive CRL nodes. CONCLUSIONS: Complete lymphadenectomy is necessary in esophageal cancer to appropriately stage patients. Low rates of positive CRL nodes are present with early clinical stage, with pT0-2 tumors, and with pN0 classification, particularly in patients with adenocarcinoma and gastroesophageal junction tumors. Dissection of the CRL field should be considered with advanced disease for adenocarcinoma and in all patients with squamous cell cancer.
BACKGROUND: Although patients with esophageal cancer (EC) often develop lymph node metastases in the cervical and recurrent laryngeal (CRL) distribution, lymphadenectomy in this field is rarely performed. The purpose of this study was to determine factors associated with CRL node positivity and to determine the appropriate indications to perform a "three field" lymphadenectomy. METHODS: In a retrospective review, EC patients who underwent three-field lymphadenectomy were analyzed. Predictors of positive CRL nodes were examined univariately, then selected for inclusion in a multivariate logistic regression model. RESULTS: From 1994 to 2009, 185 patients had a three-field lymphadenectomy, of whom 46 patients (24.9%) had positive CRL nodes. Final pathology stages (seventh edition) were I in 24 patients, II in 43, III in 109, and IV in 1 patient. Eight patients had a major pathologic response after induction therapy. On univariate analysis, variables significantly associated with positive CRL nodes included squamous cell histology, proximal location, advanced clinical presentation, the presence of clinical nodal disease, higher pT classification, and higher pN classification. There was no reduction in the rate of positive CRL nodes after induction chemotherapy. On multivariate analysis, higher pN classification (adjusted odds ratio 16.25, 95% confidence interval: 5.40 to 48.87; p < 0.0001) and squamous histology (adjusted odds ratio 6.04, 95% confidence interval: 2.21 to 16.56; p < 0.0001) predicted positive CRL nodes. CONCLUSIONS: Complete lymphadenectomy is necessary in esophageal cancer to appropriately stage patients. Low rates of positive CRL nodes are present with early clinical stage, with pT0-2 tumors, and with pN0 classification, particularly in patients with adenocarcinoma and gastroesophageal junction tumors. Dissection of the CRL field should be considered with advanced disease for adenocarcinoma and in all patients with squamous cell cancer.