Sepsis is a state of systemic inflammation resulting from an unregulated and dysfunctional response of the innate immune system to an infectious stimulus. The sepsis response is initiated through cell surface signaling receptors that recognize specific microbial antigens, such as lipopolysaccharide, from gram‐negative bacteria. The signaling pathway activation results in increased activity of transcription factors in the cytoplasm, in particular nuclear factor
κ
B, followed by translocation to the nucleus and promotion of transcription of several early phase (proximal) inflammatory mediators, such as tumor necrosis factor‐
α
and interleukin‐1
β
. A number of later phase (distal) mediators are upregulated, producing activation of inflammation, anti‐inflammation, coagulation, and apoptotic pathways, with the endothelial cell playing a central role. This response is intended to function at a local level to contain the pathogenic process; however, systemic activation results in a systemic inflammatory response.
Therapies targeting single mediators of the sepsis response based on linear, reductionist approaches to sepsis pathophysiology have been largely ineffectual. The sepsis response is perhaps best approached as a nonlinear, redundant complex network. Mathematical models based on this approach are being developed and hold promise for better understanding of the sepsis response and for developing therapeutic targets. Recently described models include a mechanistic model based on ordinary differential equations describing the behavior of populations of cells and mediators, and an agent‐based model in which agents represent individual cells associated with the inflammatory response. Both of these models show promise in reproducing particular components of the sepsis response. Future successes in the treatment of sepsis will depend on improvements in modeling that will enable a better understanding of this complex process.