Pancreatic excretion of ampicillin was evaluated in normal dogs and in dogs with induced pancreatitis. A 100-mg/kg ampicillin dose administered intravenously induced mean peak serum levels of 100 ,ig/ml, and a 200-mg/kg intravenous dose induced a mean peak serum level of 273 ,ug/ml. Ampicillin serum levels did not differ between the group of nornal dogs and those with pancreatitis. In normal dogs, the peak pancreatic fluid ampicillin concentration after the 100-mg/ kg dose was 0.4 ,ug/ml, and that after the 200-mg/kg dose was 2.7 ,ug/ml. In dogs with pancreatitis, the mean peak ampicillin concentration in the pancreatic fluid after the 100 mg/kg dose was 19,ug/ml, and that after the 200-mg/kg dose was 38.5 ,ug/ml. Pancreatic fluid ampicillin concentrations were therapeutic in dogs with pancreatitis and subtherapeutic in normal dogs.Pancreatitis is a disease associated with a considerable rate of infectious complications (3,7, 20). Pancreatic abscess is the most common local septic complication (2, 7), bearing a high rate of morbidity and mortality (10,18). This clinical situation has led to a number of experimental and clinical trials of prophylactic antibiotic regimens aimed at reducing the grave sequelae of pancreatitis. Ampicillin was the agent chosen for the current study because it is active against most enteric organisms responsible for pancreatic abscess and because it achieves high serum and bile concentrations (11). Since we could not find any information concerning the concentration of ampicillin in the pancreatic fluid, we undertook an investigation in which we measured the concentrations of ampicillin in the serum and pancreatic juice of normal dogs and dogs with induced pancreatitis after the intravenous administration of the drug.(The paper was presented in part at the 11th Intemational Congress of Chemotherapy, Boston, Mass., 1979.) MATERIALS AND METHODS Twelve mongrel dogs weighing 16 to 20 kg each underwent laparotomy. The accessory pancreatic duct was identified, ligated, and transected. A modified Thomas cannula was inserted in the duodenum exactly opposite the orifice of the main pancreatic duct. Another modified Thomas cannula was inserted into the most dependent part of the stomach as described by Bayer et al. (1). Dogs were allowed to recover for 3 weeks after the operation; they received regular food and water ad lib.Each dog was deprived of food, but not water, for 18 h before the experiment. The duodenal and gastric cannulas remained open during each experiment. The main pancreatic duct was cannulated with a glass cannula, and the pancreatic juice was collected through it. Secretin (Sigma S-3250) was infused at a rate of 0.03 U/kg per min to secure constant pancreatic juice flow throughout the experiment. After 15 min, blood, pancreatic juice, and gastric juice were collected. Ampicillin was then administered intravenously in fast infusion (over 10 min). Seven dogs received 200 mg/kg, and five dogs received 100 mg/kg. Blood, gastric juice, and pancreatic juice were obtained at 15, 30,...