The spleen is rarely a target of direct invasion by gastric cancers, but sometimes LNM is found in splenic hilus. For that reason, prophylactic removal of the splenic hilar lymph nodes in gastric cancer has been advocated by some investigators. However, it remains inconclusive whether prophylactic clearance of lymph node station No. 10 is associated with overall survival in patients undergoing resection of gastric cancer. Thus, the present study aims to investigate the impact of prophylactic No.10 lymph node clearance on the perioperative complications and prognosis of middle and upper third gastric cancer. This is the first analysis to compare the impact of G-A, G+S and G+SPSHD on the perioperative complications and prognosis of middle and upper third gastric cancer. Our findings indicated that prophylactic No.10 lymph node clearance was not recommended for treatment of upper and middle third gastric cancer.
To date, three prospective randomized controlled trials (RCTs) have evaluated the impact of prophylactic splenectomy on prognosis in patients with gastric cancer. The Csendes trial enrolling 187 patients in a single institution showed that splenectomy has no effect on survival after total gastrectomy (42% vs 36%). The Japanese trial enrolling 505 patients in multicenter showed no survival difference between G+S group and G-A group (75.1% vs 76.4%) in upper third gastric cancer. Another small-scale trial enrolling 79 patients reported by Toge et al. showed a slightly better 5-year survival with splenectomy, but the difference was not statistically significant either. In our present study, splenectomy is not recommended for gastric cancer patients without direct invasion of the splenic hilar lymph nodes because this procedure does not increase the survival rate, but only increase the postoperative complications.
Many studies have assessed the role of prophylactic splenectomy in gastric cancer patients with high risk factors for No.10 lymph node metastasis. Ohno et al. suggested that G+S is the optimal procedure in proximal T3 gastric cancer. However, Ito et al. reported that, in patients with pT3-4 tumors, prophylactic splenectomy has no significant survival benefit. Furthermore, studies show that there was no significant difference in recurrence rate and 5-year survival rate at stage III and IV the patients who underwent total gastrectomy with or without splenectomy [8, 14, 29, 40]. With respect to tumor location, Ohkura et al. found that prophylactic splenectomy has no significant prognostic impact compared with G-A in patients with tumor involving the greater curvature. It is worth mentioning that the inclusion criteria of those studies were not rigorous in selecting patients. As a result, the conclusions of these studies should be explained with cautious. Thus, large well-designed studies are needed to explore the role of No.10 lymph node clearance in patients with possible splenic hilar lymph node metastasis in the future.
It is a matter of debate whether the spleen should be preserved or removed in prophylactic splenic hilar lymph node dissection. Supporters of splenectomy argued that G+S could facilitate dissection of lymph nodes at the splenic hilum and along the splenic artery more radically, while others thought that G+SPSHD was quite enough for splenic hilar lymph node dissection. Moreover, splenectomy has been reported to be associated with increased morbidity and mortality rates due to the importance of the spleen as a part of the immune system[16, 41]. The Korean RCTs enrolling 207 patients showed slightly but not significantly better survival in G+S group than G+SPSHD group (54.8% vs 48.8%). Therefore, the study suggested that prophylactic lymphadenectomy with splenectomy was not justified, and spleen-preserved lymphadenectomy might be a better option for advanced upper and middle third gastric cancer patients. In our current study, G+S also has no advantages in prophylactic splenic hilar lymph node dissection compared with G+SPSHD.
Study comparing the prognosis between G-A group and G+SPSHD group was limited. A retrospective study by Yang et al. reported that there was no significant difference of 5-year survival rates between the two groups. Another study by Bian et al. also found that G+SPSHD could not improve the overall survival compared with G-A in patients with advanced proximal gastric cancer without metastasis to No. 4s lymph node. Meanwhile, G-A group had better short-term outcomes, faster recovery, and lower postoperative morbidity rates than G+SPSHD group. However, the two studies involved a few patients with gross invasion of the splenic hilum, so we excluded them in our present research. In our indirect comparison meta-analyses, the total complication rate and 5-year survival rate of G+SPSHD group did not differ significant from that of G-A group. However, according to the results of cumulative ranking probability plots, the G-A has highest probability to be optimal surgical procedure for patients with gastric cancer. What’s more, with respect to the safety, the fragile texture of the spleen and large amount of vessel branches being located at the splenic hilum may increase the risk of No. 10 lymphadenectomy.
There was only one meta-analysis which based on three RCTs evaluated the impact of splenectomy on long-term survival of patients with gastric cancer in the literature. Yang et al. concluded that splenectomy did not show a beneficial effect on survival rates compared to splenic preservation. However, in their meta-analysis study, they failed to make a distinction between the G-A group and G+SPSHD group, and classified them as spleen-preserving group.
There are several limitations in our present study. First, the investigations enrolled in our network meta-analysis were all retrospective studies which introduces a possible limitation of selection bias, detection bias, and performance of analysis bias. Second, we focused on overall survival only and did not analyze progression-free survival. This was partly due to literature limitations, as some studies did not report progression-free survival for one or both groups. Third, according to our inclusion criteria, the tumors were not strictly limited to upper third of the stomach.