The number of lymph nodes to be resected in surgery for non-pancreatic periampullary cancer remains unclear. The Surveillance, Epidemiology, and End Results (SEER) database was used to gather information from a large retrospective cohort. To define a novel, reasonable cut-off associated with survival, we stratified patients into subgroups depending on the number of resected lymph nodes. 1481 nodal-negative patients resected for periampullary cancer (excluding pancreatic ductal adenocarcinoma) were included. The median number of resected lymph nodes was ten. Median overall survival in the subgroup with less than 10 removed lymph nodes was 40 months, while median survival for patients with ≥10 lymph nodes was 97 months (p < 0.001). A significant survival benefit was seen if ≥ 16 lymph nodes were harvested (median survival, 117 months), while no further benefit was observed if more than 21 nodes were removed (median survival, >120 months). Sixteen or more resected lymph nodes are associated with improved survival in node-negative periampullary carcinoma. We propose to aim at harvesting and analyzing at least 16 lymph nodes.