Résumé
Background: The AJCC 2002 staging system recommends the study of at least 15 lymph nodes in the surgical specimen. Thisnumber varies and sometimes it is not achieved since it is influenced by surgeons, pathologists and patients and may modifythe patients staging and survival. Methods: This is an observational study of 201 patients with gastric cancer submitted tosurgery in the period of January 1997 to December 2001, and followed-up until July 2003 in the Brazilian National CancerInstitute. Data related to surgeon and pathologist productivity as well as survival were analyzed. Results: All patients that haveundergone D2 lymphadenectomy had more than 15 lymph nodes. The factors related to varied number of lymph nodes were:surgeon (31.5±9.3 to 43.8 ±16.8); surgical procedure (total gastrectomy, 41.4 ±15.5); extended resection (40.8 ±15.9);pN3 and stage IV (55.1 ±15.8). Multivariate analysis showed the surgeon as an independent variable (p=0,02). The morelymph nodes studied, the more nodal metastasis found (p=0.01), but this fact has been significant only in pT3 patients(p=0.007). The variables related to survival have been surgical procedure, resection of another organ, site of the tumor; pT andpN. The independent variables have been pT (p=0,01) and pN (p=0,004). Conclusion: The surgeons technique is the mainfactor related to number of lymph nodes in the surgical specimen and for accurate staging of pT3 patients over 30 lymph nodesarenecessary.