J. W. Kuethe1, E. F. Midura1, K. R. Kasten2, C. M. Freeman1, T. C. Rice1, C. C. Caldwell1 1University Of Cincinnati,Division Of Research,Cincinnati, OH, USA 2East Carolina University Brody School Of Medicine,Department Of Surgery,Greenville, NC, USA
Introduction: The successful early immune response to sepsis strikes a balance between microbial eradication and host tissue injury. Unsuccessful clearance often results in persistent inflammatory / immunosuppressive catabolic syndrome (PICS). Due to a fluctuating inflammatory state, a measure of immune status and an accurate model of risk stratification are critical to the effective use of immune modulating therapies. Determination of leukocyte numbers, their activation state, and cytokine levels has been proposed to stratify such patients. In mice, circulating IL-6 levels allow for risk stratification following cecal ligation and puncture (CLP). Although CLP is the gold standard for inducing sepsis in experimental murine models, lack of source control is a severe limitation when extrapolating to sepsis management in humans. Therefore, we utilized a CLP-Excision (CLP-E) model in which cecal excision, peritoneal wash and antibiotic treatment were performed following CLP. Using this more clinically relevant model, we hypothesized that leukocyte characterization and cytokine measurements, isolated at the time of source control, would allow us to predict survival.
Methods: Outbred mice were subjected to CLP (50% ligation / 20 gauge puncture). After CLP, the their abdomens were re-explored, the necrotic cecums debrided, the abdomens washed and a single intra-peritoneal dose of antibiotics administered. Survival was then monitored. The peritoneal wash was analyzed for IL-6 concentration by ELISA, and neutrophil numbers and activation by flow cytometry.
Results: Following excision, neutrophil characteristics and wash IL-6 levels were analyzed. After assessing for survival, the measurements associated with the mice that lived (LIVE) and those that died (DIE) were used to generate an ROC curve. Two ROC curves were significant in predicting survival (Table). A marker of neutrophil activation, CD11b was noted to be 67% more elevated in the LIVE group compared to the DIE group (p<0.0001). IL-6 concentration, a marker of inflammation, was observed to be increased 2.2 fold in the DIE group compared to the LIVE group (p<0.001). In a subsequent cohort, neutrophil CD11b and IL-6 accurately predicted risk of death using the appropriate ROC curve.
Conclusion: This technique predicts survival by analyzing surgical waste in a clinically relevant model. We observed that neutrophil activation was blunted in the DIE cohort, but elevated in the LIVE cohort. Based on this observation, we speculate that treatments to increase neutrophil activation in the DIE cohort would improve survival, but would only exacerbate host tissue injury in the LIVE cohort, thus demonstrating a need to determine immune status prior to considering immune modulating therapies.
