PROPHYLACTIC ANTIBIOTIC TREATMENT IS SUPERIOR TO ON-DEMAND THERAPY IN A STANDARDIZED MODEL OF SEVERE NECROTIZING PANCREATITIS

Abstract

Background: High morbidity and mortality rates in acute necrotizing pancreatitis are mainly caused by bacterial superinfection of necrotic pancreatic tissue and consecutive sepsis and multiple organ failure. Until today the benefit of early prophylactic antibiotics remains controversial. The aim of the present study was to evaluate prophylactic early antibiotic treatment versus on-demand therapy in a standardized model of acute necrotizing pancreatitis. Methods: Severe necrotizing pancreatitis was induced in Wistar rats by intraductal infusion of glycodeoxycholic acid (GDOC) solution into the biliopancreatic duct and Caerulein i.v.. Treatment groups received Meropenem either prophylactically six hours after induction of necrotizing pancreatitis (=before superinfection of necrosis) or therapeutically 24 hours after induction of necrotizing disease when pancreatic necrosis are superinfected. A control group received Ringer solution intravenously. After 72 hours pancreatic injury was evaluated by histology. Bacterial translocation was evaluated by bacterial cultures of pancreatic tissue and of mesenteric lymph nodes of either small bowel or colon. Furthermore, septic complications were evaluated by blood cultures and 72 hour survival was assessed. Results: In necrotizing pancreatitis without antibiotic treatment we found a mortality rate of 42.9%. Bacteriaemia, mainly with gramnegative enteral bacteria (including Pseudomonia) and Enterococcus, was present in 87.5%. Typical enteral flora was found in mesenteric lymph nodes of the small bowel and colon in 87.5% of these untreated rats. Therapeutic on-demand treatment with Meropenem reduced bacteriaemia to 50% and the mortality rate to 27.3% (not significant). Unlike this, prophylactic antibiotic treatment significantly reduced the rate of bacteriaemia to 12.5% (p = 0.01), the rate of pancreatic superinfection to 0% (p < 0.001) as well as the mortality rate to 0% (p = 0.05). Conclusion: Antibiotic prophylaxis but not antibiotic on-demand therapy reduces septic complications and mortality in experimental necrotizing pancreatitis. In contrast to clinical studies the experimental setting has the advantage of a standardized experimental design and disease severity. Thus the present study proofs that carbapenems should be given as early as possible in the course of severe necrotizing pancreatitis.