The role of surgery in patients with advanced gastric cancer.

Surgery remains the mainstay of treatment with curative intent for established gastric cancer. Patient selection is critical to achieving satisfactory outcomes, and involves careful assessment of both patient fitness and disease stage. Staging techniques have multiplied and become much more sensitive in recent years. Current best practice involves a combination of spiral CT scan, Endoscopic ultrasound, PET scanning and laparoscopy. Only a minority of patients progress to potentially curative surgery after staging and fitness assessment in Western centres. Conventional treatment involves a distal subtotal gastrectomy or total gastrectomy depending on the site of the lesion. Innovative techniques include the Merendino operation, and pylorus and nerve sparing gastrectomies for earlier stage disease in the proximal and distal stomach, respectively. There is evidence of nutritional and quality of life benefit in the first 2 years after surgery from formation of a gastric substitute reservoir. Laparoscopic resection is well established in Japan and is developing rapidly elsewhere but its role and outcomes are not yet well defined. Radical lymph node dissection remains controversial: randomised trial evidence of overall benefit is lacking, but expert series have produced excellent results, and there are indications of a sub-group benefit for patients with stage II and III disease. The increased morbidity and mortality associated with radical dissection appear to be largely attributable to pancreatic and splenic resection together with limitations in Unit expertise. Surgical palliation has become less important in recent years as interventional radiology and endoscopy techniques have been developed for the same purposes. Overall mortality and survival results have improved dramatically over the last 20 years, but interpretation of these figures is made difficult by major changes in staging and case selection. The chances of long-term survival are, however, clearly much greater, and those of peri-operative death much less for an individual patient accepted for surgery in 2006 than they would have been in 1986.