In January 1992, a standardized and more radical surgical approach in gastric cancer treatment and a standardized pathohistological workup of specimens was introduced at our institution. The aim of our retrospectively designed study was to compare the results of two consecutive groups of altogether 348 patients who underwent a total or subtotal gastrectomy for gastric cancer (period A: 162 patients operated between 1992 and 1996; period B: 186 patients operated between 1997 and 2000). The two groups of patients were comparable in age, sex, their general condition, and the proportion of potentially curable (R0) resections. There was a significant difference between the groups with regard to tumor site (fewer distal and more proximal stomach tumors in period B), type of operation (more total gastrectomies in period B), extent of lymphadenectomy (more D2 and D3 in period B), average number of affected lymph nodes (higher in period B), average number of examined nodes (higher in period B), and in UICC stage (less stage II in period B). UICC classification was impossible in 19 patients overall, but there were fewer non-classified patients in period B. The incidence of surgical complications (15.6% vs. 18.7%) and average hospital stay (14.72 days vs. 14.70 days) was not significantly different in the two groups. The drop in overall mortality rate (potentially curable and non-curable resections) in period B almost reached statistical significance (period A: 8%, period B: 6.5%; p = 0.57). The 5-year survival for all R0 resected patients (279) in 9 years was 31%. The median survival was significantly longer in period B (941 days) than in period A (570 days) for R0 and non-curable (R2) resections together (p=0.0132) as well as for R0 resections alone (period A: 844 days, period B: 1176 days; p = 0.0127). Comparison of the two consecutive groups of patients shows that after a defined period of systematic surgical work our results improved, reflected in the higher number of extracted lymph nodes, the higher number of total gastrectomies, and the longer median survival in the second period.