Can Superextended Lymph Node Dissection be Justified for Gastric Cancer with Pathologically Positive Para-aortic Lymph Nodes?

The prognosis of patients with gastric cancer and para-aortic lymph node (PALN) metastasis is poor. Recent Japanese randomized trials concluded that prophylactic PALN dissection is not effective for curable advanced gastric cancer. However, the value of curative resection in patients with pathologically positive PALN is not determined yet. We retrospectively identified 178 patients with pathologically positive PALN who underwent curative resection at the Cancer Institute Hospital from 1980 to 2004. Patient characteristics were analyzed and independent prognostic factors for death were identified by Cox proportional hazard model. Partial gastrectomy was the most frequently performed procedure (142 of 178). Postoperative morbidity and mortality rates were 30 and 2%, respectively, with a 5-year survival rate of 13.0%. Multivariate analysis revealed the total number of positive nodes (hazard ratio, 1.804; 95% confidence interval, 1.221-2.665) and macroscopic type (hazard ratio, 1.697; 95% confidence interval, 1.138-2.530) as independent prognostic factors, while age, sex, histology, pathological tumor depth, and degree of PALN dissection were not statistically significant. The 5-year survival rate increased to 28.6% in patients with a parts per thousand currency sign15 positive nodes and macroscopic type other than type 4. Prophylactic PALN dissection can not be justified in curable advanced gastric cancer. However, R0 resection including PALN retrieval might be beneficial in patients with pathologically positive PALN, providing patients are carefully selected and operations are performed safely.