Extent of lymph node dissection has been an area of controversy in gastric adenocarcinoma for many years. Some surgeons believe that cancer metastasizes through a stepwise progression, and an extensive lymphadenectomy is necessary to improve survival and/or cure the patient. Other physicians argue that extensive ly-mphadenectomies only add pe-rioperative morbidity and mor-tality and do not improve survival. Asian countries have been performing extended lymphadenectomies routinely for many years with promising survival data, although Western countries have not been able to reproduce those results. Much of the controversy surrounding lymphadenectomies started in the 1980s when Japanese studies reported superior survival rates matched stage for stage, compared to the United States. This was theorized to be secondary to the more extensive lymphadenectomy performed in Japan compared to the United States. 

A United Kingdom study randomized 400 patients to either a D1 or a D2 lymph node dissection. Those patients with tumors in the upper or middle third of the stomach underwent a distal pancreaticosplenectomy to obtain retropancreatic and splenic hilar nodes. While the 5-year survival rates were not statistically significant between the two groups, on multivariate analyses it was noted that those patients in the D2 group that did not undergo the distal pancreaticosplenectomy had an increased survival compared with the D1 group. A trial in the Netherlands randomized 380 gastric cancer patients to a D1 lymphadenectomy and 331 patients to a D2 lymphadenectomy. Similar to the United Kingdom study, there was not a significant difference in survival between the two groups, even when followed out to 11 years. There was a significant increase in postoperative complications in the D2 group compared with the D1 group (43 % vs. 25 %, respectively) as well as mortality (10 % vs. 4 %, respectively).

The data from these two studies suggest that a pancreaticosplenectomy performed to harvest lymph nodes seems to only add morbidity and mortality while not improving survival. One concern raised about the prior two studies was the variation in surgical technique and lack of standardization of surgeon experience. A Taiwanese study accounted for this by performing the study at a single institution with three surgeons, each of whom had completed at least 25 D3 lymph node dissections prior to the study. Patients with gastric cancer were randomized to a D1 lymph node dissection (defined as resection of perigastric lymph nodes along the lesser and greater curves of the stomach) or a D3 lymph node dissection (defined as resection of additional lymph nodes surrounding the splenic, common hepatic, left gastric arteries, nodes in the hepatoduodenal ligament, and retropancreatic lymph nodes). There was an overall 5-year survival benefit with the D3 group of 60 % compared with the D1 group of 54 %. A Japanese study evaluated a more aggressive lymph node dissection and randomized patients to a D2 dissection or a para-aortic lymph node dissection (PAND). There was no significant difference in 5-year survival between the two groups with a trend toward an increase in complications in the PAND group. Multiple studies have shown that the number of positive lymph nodes is a significant predictor of survival. Current AJCC guidelines stipulate that at least 15 lymph nodes are needed for pathologic examination to obtain adequate staging.

Laparoscopic techniques have become an integral part of surgical practice over the past several decades. For gastric cancer, multiple retrospective studies have reported the advantages of laparoscopic gastrectomy (LG) over open gastrectomy (OG). A recent meta-analysis of 15 nonrandomized comparative studies has also shown that although LG had a longer operative time than OG, it was associated with lower intraoperative blood loss, overall complication rate, fewer wound-related complications, quicker recovery of gastrointestinal motility with shorter time to first flatus and oral intake, and shorter hospital stay. A randomized prospective trial comparing laparoscopic assisted with open subtotal gastrectomy reported that LG had a significantly lower blood loss (229 ± 144 ml versus 391 ± 136 ml; P< 0.001), shorter time to resumption of oral intake (5.1 ± 0.5 days versus 7.4 ± 2 days; P< 0.001), and earlier discharge from hospital (10.3 ± 3.6 days versus 14.5 ± 4.6 days; P< 0.001).