Endotoxemia, trauma, hemorrhage, and infection all elicit an acute inflammatory response mediated by cells, cytokines, and other mediators that interact in a complex way. Attempts to modulate specific components of this cascade have largely failed as therapies, possibly because the of the complexity and redundancy of these interactions. We hypothesized that a mathematical simulation could accurately describe the time course of cell populations, cytokines, and key physiologic parameters of the acute inflammatory response.
We constructed a mathematical simulation of acute inflammation containing several of the key cellular and molecular components of this response. We calibrated this simulation across four experimental scenarios of acute inflammation in C57Bl/6 mice (3, 6 mg/kg LPS, trauma, and reversible hemorrhagic shock) by minimizing the error between data generated in vivo and predicted values of four analytes across all scenarios using a customized genetic algorithm. We thus produced a single simulation that could describe time course of analytes for all experimental scenarios. We validated this calibrated simulation in mice receiving 12mg/kg of endotoxin intraperitoneally by comparing experimental values of serum TNF, IL-6, IL-10, and NO2-/NO3- to theoretical predictions.
A sample of results is depicted for TNF (figure). The mathematical simulation predicts (solid line) the time course of measured analytes (mean±SEM) accurately over the 24 hours of the experiments and provides specific time courses of analytes not measured.CONCLUSIONS: The magnitude and dynamics of the inflammatory response to endotoxin in vivo was accurately predicted by mathematical simulation.
In silico modeling of complex systems provides a powerful method of integrating the concerted response of multiple interacting components of the acute inflammatory response. Such simulations could yield useful individual predictions and help individualize therapeutic interventions.
C.E. Lagoa, None.