Background: Success of surgical treatment for pancreatic and periampullary cancer is often limited due to locoregional recurrence and/or the development of distant metastases. Objective: The survival benefit of celiac axis infusion (CAI) and radiotherapy (RT) versus observation after resection of pancreatic or periampullary cancer was investigated. Methods: In a randomized controlled trial, 120 consecutive patients with histopathologically proven pancreatic or periampullary cancer received either adjuvant treatment consisting of intra-arterial mitoxantrone, 5-FU, leucovorin, and cisplatinum in combination with 30 x 1.8 Gy radiotherapy (group A) or no adjuvant treatment (group B). Groups were stratified for tumor type (pancreatic vs. periampullary tumors). Results: After surgery, 120 patients were randomized (59 patients in the treatment group, 61 in the observation group). The median follow-up was 17 months. No significant overall survival benefit was seen (median, 19 vs. 18 months resp.). Progressive disease was seen in 86 patients: in 37 patients in the CAI/RT group, and in 49 patients in the observation group (log-rank P < 0.02). Subgroup analysis showed significantly less liver metastases after adjuvant treatment in periampullary tumors (log-rank P < 0.03) without effect on local recurrence. Nonetheless, there was no significant effect on overall survival in these patients (log-rank P = 0.15). In patients with pancreatic cancer, CAI/RT had no significant effect on local recurrence (log-rank P = 0.12) neither on the development of liver metastases (log-rank P = 0.76) and consequently, no effect on overall survival. Conclusion: This adjuvant treatment schedule results in a prolonged time to progression. For periampullary tumors, CAI/RT induced a significant reduction in the development of liver metastases, with a possible effect on overall survival. Especially in these tumors, CAI/RT might prove beneficial in larger groups and further research is warranted.