It has not been established that extended lymph node resection is necessary for ductal adenocarcinoma of the head of the pancreas. According to the general rules for the study of pancreatic cancer, a multiinstitutional, retrospective clinical study was undertaken to investigate the efficiency of extended lymph node dissection for this malignancy. Altogether 501 patients underwent resection of the pancreas between 1991 and 1994 at 77 medical facilities; the surgical procedures, staging, lymph node dissection, curability, and survival rate were analyzed retrospectively. Eighteen of the patients died within 30 postoperative days, leaving 483 patients to be studied. The resection was curative microscopically in 94 patients, resulting in a 3-year survival of 29%. Macroscopically curative resection resulted in a 3-year survival of 14%; noncurative resection produced a 3-year survival of 6%. Although extended lymph node dissection was performed on 38 patients in stage I, 42 patients in stage II, 206 patients in stage III, and 1 patient in stage IV, there was no improvement in survival when the results were compared to those seen after standard or palliative lymph node dissection. The extent of lymph node dissection has not affected the prognosis for ductal adenocarcinoma of the head of the pancreas at any stage of the course of the disease. Excessive lymph node dissection in advanced cases does not necessarily lead to a favorable prognosis. The patients who undergo a radical operation with an adequate lymph node dissection have longer survivals.