The optimal extent of gastrectomy for middle-third gastric cancer: distal subtotal gastrectomy is superior to total gastrectomy in short-term effect without sacrificing long-term survival

Gastric cancer is a severe problem worldwide. It is the fourth most common cancer and the second leading cause of cancer death globally. According to recent reports, nearly 950,000 new cases are diagnosed each year, and 720,000 patients with gastric cancer died from gastric cancer in 2012 [1, 2]. Although gastric cancer is not a common cancer in North America or most Western European areas, the burden of gastric cancer is still very high in Eastern Asia, Central and Eastern Europe, and Latin America [3]. Specifically, more than 50% of patients arise in the Eastern Asian area.

Surgery is the mainstay in the multidisciplinary treatment for gastric cancer. Adequate surgical resection is the only potentially curative method for gastric cancer [4, 5]. Surgery for gastric cancer must ensure the complete removal of the tumor and potentially metastatic lymph nodes. Meanwhile, the intraoperative and postoperative safety and postoperative quality of life should be under consideration before surgeons determine the surgical treatment strategy. The extent of surgical resection is determined by tumor stage, location, size, histological type and some other clinicopathological characteristics. An adequate gastrectomy is defined as complete resection of the primary tumor with negative resection margins. According to the latest Japanese gastric cancer treatment guidelines published in 2016, the standard surgical procedure for tumor with clinically positive lymph nodes or tumor invading to or deeper than the muscularis propria is either distal subtotal gastrectomy or total gastrectomy [6, 7]. Distal subtotal gastrectomy could be selected when a satisfactory proximal resection margin can be achieved. As a result, for tumors located in the upper third of the stomach, proximal subtotal gastrectomy or total gastrectomy is recommended, depending on the depth of tumor invasion [8]. With regard to lower-third gastric cancer, distal subtotal gastrectomy is the optimal surgical procedure suggested by previous studies [9, 10]. Nonetheless, the extent of gastrectomy for middle-third gastric cancer remains controversial. Some studies have recommended total gastrectomy as the standard procedure because of its potential for improved long-term survival [11, 12]. Considering the better intraoperative and postoperative safety and quality of life, distal subtotal gastrectomy has been reported to be an alternatively curative treatment for middle-third gastric cancer [13].

The short-term effect and long-term prognosis of different extents of gastrectomy for middle-third gastric cancer have not been well evaluated until now. In our study, we therefore analyse the intraoperative and postoperative effects and long-term survival outcomes of patients with middle-third gastric cancer who underwent different extents of gastrectomy.

Surgery is the only potentially curative method for patients with gastric cancer. The ideal surgical resection not only achieves the curative intent but also decreases postoperative morbidity and mortality. The long-term prognosis and postoperative quality of life should both be of great concern [12, 16, 17]. Considering that distal subtotal gastrectomy is associated with a better quality of life and lower morbidity and mortality, many surgeons recommend distal subtotal gastrectomy as the optimal procedure for lower-third gastric cancer based on previous reports [18–20]. However, at the moment, there is no consensus regarding the best extent of gastrectomy for middle-third gastric cancer. The only prospective randomized trial, performed in Italy, compared surgical morbidity and long-term prognosis between distal subtotal gastrectomy and total gastrectomy for patients with gastric cancer. However, only approximately 20% of patients in that study had middle-third gastric cancer [10, 21]. The conclusions in that study might not be appropriate for patients with middle-third gastric cancer. Until now, many surgeons have preferred to perform total gastrectomy for relatively later-stage middle-third gastric cancer to ensure the curativeness of the surgery. In our study, the proportion of relatively later-stage cases is higher in the total gastrectomy group than in the distal subtotal gastrectomy group. This result is concordant with previous reports [13, 22].

In our study, the total number of harvested lymph nodes is higher in the total gastrectomy group than in the distal subtotal gastrectomy group (P < 0.001). The primary reason is that we perform standard D2 lymph node dissection for locally advanced gastric cancer. During distal subtotal gastrectomy, the D2 lymph node dissection includes group 1, 3, 4sb, 4d, 5, 6, 7, 8a, 9, 11p and 12a lymph nodes. With regard to the total gastrectomy procedure, the scope of D2 lymph node dissection should include the above-mentioned groups and add group 2, 4sa, 10 and 11d lymph nodes [8, 15, 23]. Therefore, more harvested lymph nodes in the total gastrectomy group is reasonable. Although there are more harvested lymph nodes in the total gastrectomy group than in the distal subtotal gastrectomy group, the total gastrectomy procedure is more complicated than distal subtotal gastrectomy. Therefore, total gastrectomy consumes more time and induces more postoperative complications and longer postoperative hospital stay. In our study, the median operative durations in the total gastrectomy and distal subtotal gastrectomy groups are 240 min and 180 min, respectively (P < 0.001). Similarly, postoperative hospital stay is longer in the total gastrectomy group than in the distal subtotal gastrectomy group (P = 0.001). Notably, the complication rate is higher in the total gastrectomy group than in the distal subtotal gastrectomy group. Further analysis by splitting the different complications elucidates that the rate of anastomosis leakage is higher in the total gastrectomy group than in the distal subtotal gastrectomy group. Previous studies obtained similar conclusions [21, 24]. The main explanation for this is that the oesophago-jejunal anastomosis is more fragile than the gastro-jejunal anastomosis. The fragility of the oesophago-jejunal anastomosis leads to a higher rate of anastomosis leakage.

The effect on long-term survival of the different gastrectomy procedures has received the most attention for a long time. In our study, overall survival in the distal subtotal gastrectomy group is better than that in the total gastrectomy group, and the 5-year survival rates in the distal subtotal gastrectomy and total gastrectomy groups are 65% and 47%, respectively (P < 0.001). Given the imbalance in tumor staging between the groups, we perform stage-stratified subgroup analysis to compare the survival outcomes. In the stage-stratified analysis, the 5-year survival rates are comparable between the groups. The results demonstrate that the difference in 5-year survival rate between the distal subtotal gastrectomy and total gastrectomy groups is mainly induced by the imbalance of tumor staging. This result is also concordant with previous reports [22, 25].

To determine the prognostic factors for survival, univariate and multivariate analyses are performed. The results show that N stage (P < 0.001), tumor size (P = 0.012), presence of lymphovascular invasion (P = 0.043), operation duration (P < 0.001), complications (P = 0.037), age (P = 0.046) and usage of neoadjuvant chemotherapy (P < 0.001) are independent prognostic factors for survival. N stage, tumor size, presence of lymphovascular invasion and usage of neoadjuvant chemotherapy all directly or indirectly represent the degree of tumor progression and TNM staging. Thus, the result that these factors could affect survival outcomes is reasonable. In our study, complication is an independent prognostic factor for survival, as also reported in previous reports [10, 25]. Interestingly, the proximal resection margin and extent of gastrectomy are not prognostic factors for survival. In fact, many surgeons prefer to perform total gastrectomy for patients with middle-third gastric cancer to make sure that they achieve a safe proximal resection margin. In previous studies as well as in our study, only if an R0 resection is achieved would the distance from the resection margin to the tumor border not affect survival outcome [22, 26]. Meanwhile, the extent of gastrectomy is also not an independent factor for survival. In other words, distal subtotal gastrectomy would be a rational choice for patients with middle-third gastric cancer if an R0 resection could be achieved.

Some limitations exist in our study. This study is a retrospective study, and selection bias is difficult to avoid. For example, the proportions of patients with larger tumor size, lymphovascular invasion, neoadjuvant and adjuvant chemotherapy or later-stage disease are higher in the total gastrectomy group than in the distal subtotal gastrectomy group. These factors all represent a likely later-stage disease. The choice of the gastrectomy procedure’s extent is decided by surgeons, who usually choose total gastrectomy for patients with later-stage disease, as total gastrectomy seems to be a more effective and safe means to achieve a curative resection. However, stage-stratified subgroup analysis and multivariate analysis are performed in our study. The results of this study remain convincing. In the future, a prospective randomised controlled study is needed to further clarify the best extent of gastrectomy for patients with middle-third gastric cancer.

For patients with middle-third gastric cancer, distal subtotal gastrectomy shortens operation duration and postoperative hospital stay and reduces postoperative complications. At the same time, the long-term survival of patients with distal subtotal gastrectomy is similar to that of those with total gastrectomy at the same stage. The extent of gastrectomy for middle-third gastric cancer is not an independent prognostic factor for survival.