Surgical treatment of distal gastric cancer
Surgical management of gastric cancer in all potentially curable stages aims to completely resect the tumor with its draining lymphatics and, once neighboring organs have been invaded, to extend the procedure to extended en bloc resection, if an R0 situation can be achieved.
While for a long time gastrectomy was considered the gold standard in the management of gastric cancer, randomized trials have demonstrated beyond doubt that total gastrectomy in gastric cancer of the distal third does not offer any benefits over partial gastric R0 resection (3, 8). Based on the recommendations of the Japanese Gastric Cancer Association (see guidelines) and the German Gastric Cancer Study 1992 (15), [MM1] the location of the tumor is the most important factor governing the extent of resection:
- Cancer in the distal third of the stomach → subtotal (four-fifths) resection
- Cancer in the middle third → gastrectomy
- Cancer in the proximal third with cardiac involvement → extended gastrectomy with transhiatal segmental resection of distal esophagus
Additional indications for total gastrectomy include linitis plastica, a special variant of gastric cancer resulting in diffuse infiltration of the gastric wall (mostly signet ring cell carcinoma) and cancer affecting several anatomical regions (upper, middle and lower third of the stomach).
Partial distal resection versus gastrectomy in gastric cancer
While in early intestinal type gastric cancer tumor spread usually is evident by gross delineation, in advanced stages, particularly in the diffuse type, discontinuous growth may be present, thereby feigning a grossly inconspicuous gastric wall. To achieve R0 resection, the diffuse type mandates an in situ oral margin of 8 cm. with 5 cm in the intestinal type. With this in mind, partial distal gastric resection is indicated for early gastric cancer of both types as well as for the intestinal type in advanced stages with tumor sites in the middle and distal third. If the histological classification of the tumor type is known and the recommended oral margins are observed, the extent of resection may be planned before surgery (9, 13, 14).
In partial distal gastric resection, about 80% of the distal stomach must be resected to achieve freedom of tumor at the oral resection line. At the lesser curvature the oral resection line should be about 2 cm distal to the cardia and at the greater curvature proximal to the union of the right and left gastroepiploic vessels; while the distal resection line should be about 3 cm distal to the pylorus. The procedure requires not only gastric resection and lymphadenectomy, but also resection of the lesser omentum close to the liver and subtotal resection of the greater omentum. During gastrectomy the latter is resected in toto (14, 17).
Cancer in the proximal third of the stomach requires extended transhiatal gastrectomy. A transthoracic approach is only indicated if the transhiatal approach does not achieve tumor freedom at the distal esophageal resection line, which is easily verified by frozen section. A primarily transthoracic procedure is not recommended (13, 17).
Studies have shown that in gastric cancer of the middle and distal third both partial distal gastric resection and gastrectomy result in identical outcome. Postoperative morbidity and mortality do not differ significantly; if oral margins are respected, the five-year survival rates do not differ. However, partial resection generally offers a better quality of life than gastrectomy (3, 8, 17).
At present, the various resection procedures are mostly performed as open surgery. Although the number of resections performed laparoscopically, especially in early gastric cancer, is rising steadily, final assessment of the value of minimally invasive techniques is not possible at present due to the lack of randomized trials with larger case numbers and long-term outcomes (2, 10, 18).
Reconstruction of the intestinal passage
After partial distal gastric resection, intestinal passage is generally restored by end-to-side gastrojejunostomy. In order to spare patients the burden of biliary reflux with reflux gastritis (in gastrectomy: esophagitis), Roux-en-Y reconstruction with a jejunal limb of at least 40 cm between oral anastomosis and jejunojejunostomy is recommended. A Braun jejunojejunostomy does not completely prevent alkaline reflux.
Lymphadenectomy ("LAD")
Apart from an R0 resection, the extent of the lymphatic tumor spread is an important prognostic factor in gastric cancer, and therefore the draining lymphatics in gastric cancer should also be resected with adequate margin to further improve the prognosis of the patient, as demonstrated by the removal of enough lymph nodes without cancer spread. If less than 20 % of the distant lymph nodes are invaded, the prognosis is significantly better than with a less favorable lymph node involvement (20). It is therefore important to clear as many lymph nodes as possible without increasing morbidity and mortality, which otherwise would endanger the benefits of LAD.
Three prospective randomized European trials studied the radicality of lymph node dissection in gastric cancer, comparing LAD in compartment 1 (perigastric lymph nodes) with extended LAD (compartments 1 and 2 = lymph nodes along left gastric artery, celiac trunk, common hepatic artery, splenic artery. and hepatoduodenal ligament), the so-called D2-LAD (1, 6, 7).
In the Dutch (1) and English (6) trials, morbidity and mortality increased significantly after extended LAD, however, in both trials the percentage of patients undergoing splenectomy and/or pancreatic resection was significantly higher. A closer look at the two trials reveals that increased morbidity and mortality was primarily due to complications resulting from pancreatic surgery and, in particular, splenectomy. The third D2-LAD trial (7) considered spleen and pancreas sparing procedures; it did not find any significant difference in morbidity and mortality compared to D1-LAD. A Cochrane review from 2004 underlines the improved staging accuracy and prognosis in D2-LAD (12).
A randomized Japanese clinical trial studied whether extending the D2-LAD by additional dissection of paraaortic lymph nodes offers additional benefits (16). Compared to D2-LAD, morbidity and mortality increased only slightly, but no significant difference in prognosis was observed. Thus, additional extension of the D2-LAD cannot be recommended at present.
Conclusion: Compared to D1-LAD, subtle dissection in D2-LAD does not increase morbidity and mortality. D2-LAD improves the prognosis in gastric cancer as long as splenectomy (with/without resection of the pancreatic tail) is consistently avoided.
Splenectomy and resection of the pancreatic tail
Routine splenectomy should not be performed in gastric cancer and is also not justified in gastrectomy (5). Distal pancreatic resection with/without splenectomy may be indicated in stage T4, but only if this permits R0 resection (11).
Endoscopic resection of early gastric cancer
Superficial early cancer limited to the gastric mucosa (T1aN0M0) may be resected endoscopically because the risk of lymphatic spread is 0 – 2 %. If, on the other hand, the superficial region of the submucosa has already been invaded, the rate of lymph node metastasis jumps to 25 %. According to the Japanese classification of gastric cancer, endoscopic submucosal resection may be performed in mucosal cancer, if:
- Lesion diameter < 2 cm, elevated type
- Lesion diameter < 1 cm, flat type
- No ulceration
- Histology: well to moderately differentiated (G1-G2)
The aim is the en-bloc-R0-resection respecting the margins required by the histology of the tumor (intestinal type: 4 – 5 cm, diffuse type: 5 – 8 cm). Patients with helicobacter pylori colonization should be eradicated first.
Palliation
For symptomatic stenosis, bleeding and metastasis of the tumor, the current German S3 guideline on gastric cancer surgery offers these consensus statements based on published trials:
Tumor stenosis
- Stenting, gastroenterostomy, jejunal feeding tube or palliative radiotherapy
- Palliative resection: Only in exceptional cases, since there is insufficient evidence
Tumor bleeding
- Endoscopic hemostasis; if not possible or ineffective:
- Angiographic embolization
- Palliative resection as a measure of last resort
- Chronic seepage: Palliative radiotherapy
Metastases
Regarding survival, there is insufficient evidence at present for the effectiveness of surgery in metastatic spread. In individual cases, resection of isolated organ metastases (liver, ovaries) may be considered in the absence of peritoneal spread.