**4. Clinical difference of lymph node metastasis between ESCC and EAC**

#### **4.1. Lymph node metastasis of ESCC**

Udagawa et al. showed that it was necessary to dissect the cervical lymph nodes, particularly for tumors located in the upper and middle thoracic esophagus [2]. Bilateral cervical paraesophageal node dissection is predominantly effective and the inclusion of these stations in the regional lymph nodes is justified in the 7th TNM classification. The lower jugular or supraclavicular region lymph nodes are also important. Abdominal lymph node dissection is also effective, but the effectiveness is limited in upper esophageal cancer. The statistical efficacy index (EI) of individual lymph node stations according to the main tumor location, a more precise modification of the range of lymph node dissection is possible. Although mediastinal nodes are important in general, not all are located in the esophageal drainage area. Some specific stations such as pretracheal, left tracheobronchial, and supradiaphragmatic show least efficacy by dissection. The mediastinal node stations can be re-arranged such as recurrent laryngeal, paraesophageal, posterior mediastinal, subcarinal, and subbronchial in order, according to EI. The number of metastatic lymph nodes may be a better prognostic factor than Japanese N grading. The Japanese N-grouping seems to be more efficient for predicting of the radical and safe operation. It is difficult to detect lymph node metastasis clinically. Japanese N-grouping may be also more available in stage assessment because a single obvious distant lymph node metastasis in Japanese N-group can correctly identify a higher staging. Japanese Surgeons are afraid that discussion about lymph node dissection with precision is no longer possible if meticulous node grouping based on detailed data of lymph node stations is once discontinued.

Ma et al. stated that 3FL improves overall survival rate but has more complications. Because of the high heterogeneity among outcomes, definite conclusions are difficult to draw [16].

#### **4.2. Lymph node metastasis of EAC**

**3. Clinical difference between EAC and ESCC**

third of the esophagus, such as studies conducted in Ghana [15].

In 50–60% of cases, ESCC is located in the middle third thoracic esophagus in Thailand, Iran, and Japan [3, 12, 13, 14]. About 60% of SCCs is also located in the middle third of the esophagus in Iran. Several studies have found that ESCC was most commonly located in the lower

In 80% of cases, EAC is located at the gastroesophageal junction (GEJ) and 20% in the lower third thoracic esophagus. This entity is capable of producing EAC directly or, more commonly, through an intermediate pre-neoplastic lesion or Barrett's esophagus (BE). BE is a pre-malignant lesion that develops in 6–14% of patients with Gastroesohageal reflux disease (GERD), of which approximately 0.5–1% will develop EAC [10, 11]. Increased incidence of BE in the past 30 years correlates with an increased incidence of EAC during the same

**4. Clinical difference of lymph node metastasis between ESCC and EAC**

Udagawa et al. showed that it was necessary to dissect the cervical lymph nodes, particularly for tumors located in the upper and middle thoracic esophagus [2]. Bilateral cervical paraesophageal node dissection is predominantly effective and the inclusion of these stations in the regional lymph nodes is justified in the 7th TNM classification. The lower jugular or supraclavicular region lymph nodes are also important. Abdominal lymph node dissection is also effective, but the effectiveness is limited in upper esophageal cancer. The statistical efficacy index (EI) of individual lymph node stations according to the main tumor location, a more precise modification of the range of lymph node dissection is possible. Although mediastinal nodes are important in general, not all are located in the esophageal drainage area. Some specific stations such as pretracheal, left tracheobronchial, and supradiaphragmatic show least efficacy by dissection. The mediastinal node stations can be re-arranged such as recurrent laryngeal, paraesophageal, posterior mediastinal, subcarinal, and subbronchial in order, according to EI. The number of metastatic lymph nodes may be a better prognostic factor than Japanese N grading. The Japanese N-grouping seems to be more efficient for predicting of the radical and safe operation. It is difficult to detect lymph node metastasis clinically. Japanese N-grouping may be also more available in stage assessment because a single obvious distant lymph node metastasis in Japanese N-group can correctly identify a higher staging. Japanese Surgeons are afraid that discussion about lymph node dissection with precision is no longer possible if meticulous node grouping based on detailed data of lymph node

**3.1. Location of ESCC and EAC**

76 Esophageal Abnormalities

**4.1. Lymph node metastasis of ESCC**

stations is once discontinued.

period.

Sepesi et al. reported 72% of patients presented with clinically involved lymph nodes showed metastasis in the lesser sac (perigastric/perihepatic) [17]. However, 11% of patients had metastatic lymph nodes located at the celiac artery, and 10% had nodal disease in the paratracheal region. Ninety-eight patients demonstrated clinical metastatic involvement in one or two nodal basins (example: perigastric and paratracheal); only about 2% of patients presented nodal disease in three nodal basins. Feith et al. reported the prevalence and number of lymph node metastases according to pT category in patients with primary resected Barrett's carcinoma. A strong correlation between the pT category and the presence and the number of lymph node metastases was detected [18]. Lymph node metastases in more than 95%of the patients was detected in the lower posterior mediastinum, in the bilateral paracardiac region, or in the region of lesser curvature and left gastric artery of the abdomen. Prevalence of lymph node metastases at the various topographic locations in relation to the T category of the underlying Barrett's carcinoma is shown. As T category increases, prevalence of regional lymph node metastases also markedly increases. Lymph node metastases to more distant locations, such as the tracheal bifurcation region, the proximal mediastinum, or celiac axis, lagged behind and were common only in patients with more advanced primary tumors. This suggests that lymphatic spread occurs in an orderly fashion. Patients who had many lymph nodes metastasis had distant lymph node metastasis in the upper mediastinum and tracheal bifurcation area that can be reached by a transthoracic approach [19]. According to an increasing number of lymph nodes metastasis, positive lymph node metastasis in the upper mediastinum and carinal region also increased. Prevalence of lymph node metastases at the celiac axis was also detected with increasing number of lymph nodes metastasis. There was significantly frequent lymph nodes metastasis in the upper mediastinal, carinal or celiac regions among patients with more than three positive regional nodes. Overall, skipping of regional lymph node stations, positive distant nodes in the absence of positive regional nodes, was seen in less than 5% of the patients.

Yamashita et al. reported clinical records of 2807 EGJ carcinoma patients without preoperative therapy in Japan [20]. There are obvious unbalances in terms of lymph node dissection rate according to histology and the main tumor location. Lymph nodes metastasis frequently involved abdominal lymph nodes. Lymph nodes at the right and left cardia, lesser curvature and along the left gastric artery were especially metastasized. Lymph nodes along the greater curvature of the stomach were not frequently metastatic, and advantage of dissection seemed unlikely. Lower mediastinal node dissection may contribute to improved survival for patients with esophagus-predominant EGJ carcinoma. However, due to low dissection rates for nodes of the middle and upper mediastinum, no conclusive results have been obtained regarding the optimal extent of nodal dissection in this region.
