Patients with solitary lymph node metastasis in esophageal cancer has been considered a distinct prognostic subgroup with cancer outcomes closer to node-negative disease than any other node-positive subgroup. It has even been believed that there is no difference in survival between patients with solitary lymph node metastasis and N0 patients with esophageal squamous cell carcinoma, and that solitary lymph node metastasis does not affect the prognosis. In this report, except for T category, all clinicopathological variables did not differ between N0 patients and patients with solitary lymph node metastasis. Solitary lymph node metastasis had a worse survival compared with N0 disease. These results may partly explained by the fact that a majority of patients with solitary lymph node metastasis were at the T3-stage. But skip metastasis was not a risk factor for poor prognosis. According to the JSED criterion both skip (+) and skip (−) patients showed worse survival than N0 patients, but it was T category not N status that was the independent prognostic factor in multivariate analysis. Although N status was the independent prognostic factor according to the other two criteria, there was still no significant difference between N0 and skip metastasis in predicting prognosis.
The prognostic impact of lymph node skip metastasis in esophageal carcinoma has been unclear. Some reports have shown a favorable prognosis of skip metastasis (compared with adjacent or continuous lymph node metastasis) and others have not[7, 18]. Prenzel et al. attributed the different results to the different lymph node classification used to define skip metastasis. In this series, three criteria were used to define skip metastasis. Skip metastasis did not show a favorable prognosis compared with adjacent metastasis. But multivariate analysis did show that skip metastasis was not a prognostic factor, whereas adjacent metastasis was prognostic in two of the three models. It must be noted that skip metastasis could be caused by inefficient histopathologic examination by missing out small metastasis in positive nodes. In addition, micrometastasis detected by immunohistochemistry in histologically tumor-free nodes is not a rare event in esophageal cancer[6, 18]. Natsugoe et al. reported micrometastasis was found in 33 of 59 patients (55.9%) with esophageal carcinoma, who were diagnosed as N0 by routine histological examination.
As previous reports have shown[4–6, 8], solitary lymph node metastasis in this report was widely distributed, but was mainly limited to the recurrent nerve region and middle thoracic paraesophagus region. It was reasonable that the JSED classified these two regions as N1 in middle thoracic squamous cell carcinoma. Some authors have found that solitary lymph node metastasis in esophageal carcinoma is predominately located in the abdomen along the lesser curvature and the left gastric artery. But their series included lower thoracic tumors and upper mediastinal lymphadenectomy was not routinely performed. The wide distribution of solitary lymph node metastasis was closely associated with the tumor site and unique features of lymphatic spread from the esophagus, which has been discussed in many previous reports[4, 5, 8, 19, 20]. In addition, middle thoracic tumors have a more obvious tendency to develop bidirectional metastasis than upper and lower thoracic tumors of the esophagus[1, 19, 20], so it is still hard to predict the location of solitary lymph node metastasis, even if sentinel navigation surgery is used.
Three-field might be more appropriate than two-field lymphadenectomy for middle thoracic esophageal squamous cell carcinoma in terms of the wide distribution of solitary lymph node metastasis. But there was no survival difference between patients treated with these two types of lymphadenectomies (data not shown). Many similar results reported earlier have also been summarized in reviews of the literature[21, 22]. So far, controversy exists as to the extent of lymphadenectomy required in esophageal cancer. There is no doubt that more extensive lymphadenectomy provides more accurate nodal staging, but whether it improves survival still needs to be tested in well-designed randomized controlled trials.
Several potential shortcomings in this retrospective study deserve attention. Surgeons chose different types of lymphadenectomy depending on their preferences. A certain selection bias can therefore not be excluded. It is also possible that variation in the quality of lymphadenectomy among different surgeons may interfere with the results. In addition, the sample size was limited by nature of the single institution of the study. Validation from other institutions is needed. Nonetheless, this series proved to be homogenous with regard to many clinical variables, such as tumor site, pathological type, and preoperative management.