BACKGROUND: Patients with lymph node-negative gastric cancer show a better overall survival rate than those who have a pathological lymph node-positive gastric cancer. But a large number of patients still develop recurrence. We aimed to explore the significant prognostic factors of lymph node-negative gastric cancer and determine how many lymph nodes should be removed.

METHODS: A total of 3103 patients who underwent radical operation are identified from the Surveillance, Epidemiology, and End Results database. Standard survival methods and restricted multivariable Cox regression models were applied.

RESULTS: The overall survival rate was significantly higher with an increasing number of negative lymph node resected. Among the 843 patients who had the exact T stage, the overall survival rate was significantly better in T3-4 group with more than 15 lymph nodes resected (P < 0.001) but not in T1-2 stage patients (P = 0.44). A further 25 more lymph nodes resection did not show additional survival benefits. Multivariate analysis of patients demonstrated that age, depth of tumor invasion, and the number of lymph nodes resected were the significant and independent prognostic factors.

CONCLUSIONS: A lymphadenectomy with more than 15 lymph nodes removal should be performed for T3-4 lymph node-negative gastric cancer. But the survival benefit of a lymphadenectomy with more than 25 lymph nodes removal is disputed. And the further treatment should refer to the prognostic indicators.