Despite encouraging preclinical findings, clinical experience with immunomodulatory therapies in patients with sepsis have been disappointing. Unlike preclinical sepsis models, patients with sepsis may not be a uniform group, however, and important factors that differ between patients may influence the effects of immunomodulators. Although many of the immunomodulators tested clinically thus far have been anti- inflammatory agents, interest is developing in the use of pro-inflammatory agents to stimulate host defense in patients with infection and sepsis. One such agent is granulocyte colony stimulating factor (G-CSF). In order to investigate the factors that might alter the effects of G-CSF, we developed a small animal model of infection and sepsis. Using a multifactorial design in this antibiotic treated rat E. coli sepsis model, we have now demonstrated that site and severity of infection fundamentally alter the survival effects of pretreatment with G-CSF. With intravenous E. coli challenge, G-CSF pretreatment increased lethality at LD10, LD50 and LD90 bacterial dosages. However, with intrabronchial challenge, although G-CSF worsened outcome at LD10 and LD50 bacterial dosages, it improved outcome at LD90 dosages. These findings have been confirmed in subsequent studies. Investigations with this model are now being continued to determine: 1) Whether G-CSF pretreatment will have differential survival effects with increasing dosages of E. coli administered intraperitoneally. 2) The mechanisms underlying the contrasting effects of G-CSF on survival with increasing intrabronchial bacterial challenge. This set of studies will determine whether identifiable markers of infection or inflammatory tissue injury correlate with either the protective or detrimental effects of G-CSF. 3) Whether G-CSF pretreatment will have differential effects with changing strains of bacteria (E. coli vs. Klebsiella vs Staphylococcus).
