Background: Endoscopic ultrasound (EUS) is known to be accurate for staging pancreatic cancer. Little data exist to determine if linear array or radial scanning EUS is superior for staging pancreatic cancer. This prospective comparison was undertaken to assess the accuracy of linear array and radial scanning EUS for staging pancreatic cancer. Methods: Patients with pancreatic cancer referred for EUS staging were randomized to linear array or radial scanning EUS. Staging accuracy for each was determined by comparison to surgical pathology in those patients going to surgery. Results: Seventy-nine patients with pancreatic cancer were enrolled and 33 had surgical resection. Of these, 17 patients were randomized to linear array and 16 to radial scanning EUS. The remaining 46 patients did not have surgery because of comorbid illness or clinically unresectable disease. EUS staging accuracy for linear array was 94% (16 of 17) for T and 71% (12 of 17) for N staging, whereas radial scanning was 88% (14 of 16) for T and 75% (12 of 16) for N staging. For predicting vascular invasion, radial scanning was 100% accurate (16 of 16) while linear array was 94% (16 of 17) accurate. There was one false-negative assessment of invasion using linear array EUS. Conclusion: Overall, both EUS designs appear equivalent for staging pancreatic cancer and assessing vascular invasion. In view of our findings and the capability for ultrasound-directed fine-needle aspiration with linear array EUS, this instrument may be the preferred choice for evaluating pancreatic masses. (Gastrointest Endosc 1997;45:138-42.)