Gastric emptying in patients after several upper gastrointestinal surgeries was studied using the acetaminophen method. The subjects consisted of 23 gastric cancer patients, 2 duodenal ulcer patients, 5 periampullary cancer patients and 4 normal subjects. As an indicator of the gastric emptying rate, the serum acetaminophen concentration was measured by fluorescence polarization immunoassay, in units of g/ml, at 0, 30, 60, 120, and 180 minutes after ingestion of a liquid meal with 1.5 g of acetaminophen. In the normal subjects, the acetaminophen concentrations were 0, 16.35+/-5.06, 18.71+/-5.58, 16.38+/-4.82, and 11.09+/-3.62 g/ml at time 0, 30, 60, 120, and 180 min, respectively. The concentration peaked at 60 min after ingestion of the test meal in the normal subjects. We observed significant delayed gastric emptying after pancreas preserving pancreaticoduodenectomy (PPPD) and a standard Whipple's operation in the early postoperative period. In all patients with a subtotal gastrectomy, a truncal vagotomy was done. However, in patients with a pancreaticoduodenectomy, the vagus nerves were preserved. The gastric emptying pattern was different between the patients with a subtotal gastrectomy and the patients with a pancreaticoduodenectomy, despite similar reconstructions of the gastroenterostomy (Billroth I or Billroth II type reconstruction). There was more rapid gastric emptying in patients with a truncal vagotomy and pyloroplasty than in the normal subjects. Hence, we speculate that the truncal vagotomy was the main cause of the different gastric emptying between the patients with a pancreaticoduodenectomy and the patients with a subtotal gastrectomy.