RÉSUMÉ
<p><b>OBJECTIVE</b>To determine the minimum number of lymph nodes that should be examined for the UICC/AJCC TNM classification of gastric cancer.</p><p><b>METHODS</b>The clinical and pathological data of four hundred and thirty-six patients underwent curative resection for gastric cancer were analyzed by Chi-square and Student-Newman-Keuls test.</p><p><b>RESULTS</b>The pN0 patients with 1 to 4, 5 to 9 examined nodes showed significantly lower survival rate than those with 10 to 14, 15 or more examined nodes (P < 0.05), and the patients with 10 to 14 examined nodes had as good a prognosis as those with 15 or more examined nodes. In the pN1, pN2 categories, the patients with 1 to 4, 5 to 9 and 10 to 14 examined nodes tended toward significantly lower survival rates than those with 15 or more examined nodes (P < 0.05). Among the patients who were classified as stage II, the survival rate of those with 10 to 19 examined nodes was significantly lower than that with 20 or more examined nodes. Among the patients classified as stage III, those with 5 to 9, 10 to 19 and 20 to 29 examined nodes had significantly lower survival rates than those with 30 or more examined nodes (P < 0.05).</p><p><b>CONCLUSIONS</b>The number of lymph nodes examined has significant prognostic impact within each pN category of gastric carcinoma. The minimum number of lymph nodes to examine in order to determine pN0 classification can be reduced from 15 to 10. For tumor with lymph node metastasis, the minimum number should be 15. In stage II, 20 or more nodes should be examined, and in stage III and IV 30 or more.</p>