OBJECTIVE To test efficacy of murine monoclonal, rabbit polyclonal recombinant equine or human tumor necrosis factor-alpha (rETNF or rHTNF, respectively) antibodies to inhibit native equine tumor necrosis factor (TNF) activity. METHODS 8 and 18 healthy adult horses for parts 1 and 2 of the study, respectively. METHODS In part 1, supernates from endotoxin-activated peritoneal macrophages were incubated with various dilutions of each rETNF antibody and subsequently tested for TNF activity. Serum was also obtained from a horse 1 hour after infusion with 20 ng of endotoxin/kg of body weight and was incubated with various dilutions of rabbit polyclonal rHTNF antibody. In part 2, 20 ng of endotoxin/kg was infused in horses during a 30-minute period. Fifteen minutes after the endotoxin infusion was initiated, 1 of 3 preparations was infused: 0.1 mg of rabbit polyclonal (rHTNF antibody/kg, 0.1 mg of human IgG/kg, or 500 ml of 5% dextrose. Clinical and hematologic data were collected for 24 hours. RESULTS Compared with the monoclonal antibody, the rabbit polyclonal rETNF antibody was more effective in inhibiting TNF activity. The 50% effective doses of the murine monoclonal rETNF, rabbit polyclonal rETNF, and rabbit rHTNF antibodies were 1.8, 0.8, and 0.6 micrograms of antibody/ml, respectively. In part 2, endotoxin infusion resulted in significant alternations in all variables; however, differences among treatment groups were not significant. CONCLUSIONS Although murine monoclonal and rabbit polyclonal rETNF or rHTNF antibodies are capable of inhibiting native equine TNF activity in vitro, when given after initiation of endotoxemia, administration of 0.1 mg of rabbit polyclonal rHTNF/kg does not alter the response to infusion of endotoxin.