Should we remove splenic hilus lymph nodes for esophagogastric junction adenocarcinoma?

Abstract

For the surgical treatment of gastric cancer with curative intent, two issues are of main importance: the radical removal of the tumor and the radical removal of affected lymph nodes. For radical removal of the tumor, margins up to 8 cm for diffuse-type carcinoma (German guidelines) are advised because microscopic positive resection margins lead to poor prognosis [1]. In a comparative study of 619 patients, it was shown that intraoperative re-excision of microscopically involved resection margins may be beneficial for patients with less than five affected lymph nodes [2]. When Billroth performed his first gastric resection in 1881, he actually also removed some enlarged nodes. His patient died 14 months after the operation of recurrent disease. Since then, the extent of lymph node dissection has been an issue in the treatment of gastric cancer. Theoretically, removal of a wide range of lymph nodes improves the chances for cure. Such resection, however, could be irrelevant when no lymph nodes are affected, or when the cancer has developed into systemic disease, or the dissection increases morbidity and mortality substantially. From randomized studies, nowadays a D2 lymph node dissection is advised [3]. More extended lymph node dissections did not lead to improved survival, as shown by Sasako et al. [4], in a randomized D2 versus D4 trial in Japan. One of the main problems in the Western D1-D2 trials was the occurrence of major complications in the D2 dissection group, mostly caused by pancreatico-splenectomy. The increased morbidity and mortality in the D2 group probably offset the survival gain at that time. Subgroup analysis from the Dutch gastric cancer trial of patients who did not undergo a pancreatico-splenectomy showed a significant survival advantage for those who had a D2 lymph node dissection (11-year survival, 33 % for D1 and 47 % for D2; p = 0.018) [3]. In a late follow-up study it was shown that local recurrence rate and cancer-related death were significantly higher in the D1 dissection group as compared to the D2 dissection group [5]. To make lymph node dissections more ‘tailor made,’ Prof. Maruyama introduced the Maruyama index of unresected disease [6]. Based on seven input variables (age, sex, Borrman type, tumor size, tumor position, and histology), the likelihood for nodal involvement can be calculated. The chance of affected lymph nodes along the splenic artery (station no. 11) and the splenic hilum (no. 10) is highest is patients with large, proximal, poorly differentiated tumors. For adequate removal of these lymph node stations, a splenectomy is often performed. Another way to individualize gastric cancer treatment is the use of the index of estimated benefit from lymph node dissection (IEBLD) [7]. This index is calculated by multiplying the frequency of lymph node metastasis to each station by the 5-year survival rate of patients with positive lymph nodes at each station. To evaluate the effect of splenectomy for gastric cancer, three randomized trials have been performed in the past. In these randomized trials from Chile [8], Korea [9], and Japan [10], researchers reported no survival benefit from splenectomy in patients with a total gastrectomy, whereas morbidity was increased. A meta-analysis of operative survival with the data from these randomized trials showed no statistically significant difference (OR, 1.59; 95 % CI, This editorial refers to the article at doi:10.1007/s10120-013-0214-y.