A low-morbidity murine model of peritonitis.

Abstract

Peritonitis continues to be a major source of mortality and morbidity in patients undergoing abdominal surgery. The aim of this study was to develop a nonfatal model of bacterial peritonitis in mice so that we could study aspects of the pathobiology and treatment of peritonitis in an in vivo model. Mice were inoculated via a midline laparotomy with 0.5 mg of zymosan in 0.1 ml of saline into the subomental space. At 24, 48, and 96 hours after treatment, animals were killed, and analysis was performed to determine the degree of peritoneal inflammation. End points included intraperitoneal cellular influx, tumor necrosis factor-alpha concentrations, and myeloperoxidase activity. In addition, peritoneal lavage fluid was plated onto blood agar for analysis of bacterial colony-forming units. There were 40 mice in each group. There were no deaths in either group. Facultative Gram-negative bacteria were cultured from the peritoneal cavities of zymosan-treated animals at 24 and 48 hours after insult (colony-forming unit counts of 92+/-11 vs. 0 in control animals). In the zymosan-treated animals, there were significantly increased numbers of inflammatory cells (especially neutrophils) in the peritoneal cavity at 24 and 48 hours after treatment; these numbers returned to control levels by 96 hours. Myeloperoxidase activity was also elevated both in the peritoneal fluid (2.4 X 10(-4) units/ml compared with 1.6 x 10(-4) units/ml, P < 0.05) and in remote organs (i.e., lung, P < 0.05; liver, P < 0.001; and kidney, P < 0.05) at 24 hours after treatment. There was no significant difference in tumor necrosis factor-alpha levels between zymosan-treated and control mice in either serum or peritoneal fluid at any time point investigated. There was no mortality in either the zymosan-treated or control animals. In this model of bacterial peritonitis in mice, we have demonstrated how the peritoneal cavity can resolve a relatively localized inflammatory insult within 96 hours of induction. This response is characterized by a cellular influx of predominantly neutrophils and macrophages and by pronounced oxidative activity. We will use this in vivo model to characterize aspects of the pathobiology and treatment of peritonitis.