Background and Objective Esophageal cancer is a highly malignant tumor with a poor prognosis.Lymph node involvement and the number of positive nodes,especially the cervical and recurrent laryngeal nerve nodes,are independent prognostic predictors after esophagectomy.Three-field lymphadenectomy may prevent recurrence and prolong survival for esophagus carcinoma when compared with Two-field lymphadenectomy,whereas it is also a highly invasive procedure that can lead to a higher complications.The status of cervical lymph nodes(CLN)is one of the most important factors to identify whether 3-FL should be performed.Unfortunately,current clinically used approaches demonstrated a lower sensitivity and specificity in identifying CLN metastasis.Recurrent laryngeal nerve nodes also have been regarded as the sentinel nodes to predict CLN with an accuracy of less than 50%.We will perform the present study from three direction as following.The first objective of present study is to analysis the outcomes and risk factors of CLN for thoracic squamous cell carcinoma of the esophagus,and futher to defined candidates benefit from three field lymphadenectomy.The second aim is revealling the accuracy of CLN sizes detected by ultrasonography to predict CLN involvement,and depict the standared criterion.The last one is identifying correlation between recurrent laryngeal nerve nodes and CLN metastasis,and demonstrate whether recurrent laryngeal nerve nodes should be regarded as the sentinel nodes to predict the CLN metastasis.Methods We retrospectively reviewed the medical records of patients with thoracic squamous cell carcinoma of the esophagus,who were admitted between January,2010 and January,2016 to ward one,the Department of Thoracic Surgery,Sichuan Cancer Hospital.Disease stage was presented according to the 2009 American Joint Committee on Cancer(AJCC)and Union for International Cancer Control(UICC)tumor–node–metastasis(TNM)system.Results One hundred and thirty-five patients met the inclusion criteria.The overall rate of CLN metastasis was 36.3% for all patients and was 46.9%,25.6%,and6.7% for patients with upper,middle,and lower thoracic esophageal tumors,respectively(P= 0.003).There was difference between the rate of CLN for middle esophageal tumors with or without mediastinal nodes metastasis(38.9% vs 19.8%,P=0.029).Multivariate analysis showed that the independent prognostic factors for CLN were the short axis of CLN detected by ultrosonography,left recurrent nerve nodal metastasis,abdominal nodes metastasis,pathological staging and tumor location.The sensitivity,specificity,positive and negative predictive value of the US assessment of CLN metastases with cut-off values of ≥0.5cm and ≥1cm were 75.8% vs 50.0%,64.9% vs91.8%,39.2% vs 64.6% and 90.0% vs 86.0%,respectively.Areas under the curve(AUC)of receiver operating characteristic(ROC)for the short axis of CLN was 0.768,the sensitivity and specificity of the best threshold value >0.5cm were 74.2% and 96% 74.0%.The AUC of short axis >0.5cm combined location was 0.809,the sensitivity and specificity for patients with upper,middle,and lower thoracic esophageal tumors were 71.4% vs 83.3% vs 100%and 75.2% vs71.2% vs75.9% when the cut-off value>0.5cm,respectively.The overall rate of recurrent laryngeal nerve nodes metastasis was 53.3%.The rate of CLN metastasis for all patients with or without recurrent nerve nodal involvement were 44.4% and27.0%(P=0.036).The positive predictive value,negative predictive value,sensitivity,and specificity of recurrent laryngeal nerve nodes as the sentinel nodes to predict the CLN metastasis were65.3%,53.5%,44.4% and 73.0%,respectively.Conclusions We concluded that CLN dissection should be recommended for patients with upper thoracic esophageal cancer and short axis of CLN >0.5cm detected by ultrosonography;The optimal diagnostic indicator of size for the ultrosonography to predict CLN metastasis is short axis,and the cut-off value is0.5cm;The recurrent nerve nodal alone should be not regarded as the sentinel node of esophageal carcinoma to receive CLN dissection or not. |