Splanchnic artery occlusion shock was induced in anesthetized rats by clamping splanchnic arteries for 45 min. Survival rate, serum and macrophage tumor necrosis factor (TNF-alpha), peritoneal macrophage phagocytosis, and killing activities were evaluated. Shocked rats died within 2 hr, whilst all sham-shocked rats survived more than 6 hr. Serum and macrophage TNF-alpha was undetectable in sham-shocked rats while shocked rats exhibited increased serum (110 +/- 5 U/ml 90 min after release of occlusion) and macrophage levels (122 +/- 4.5 U/ml 90 min after release of occlusion) of TNF-alpha. Furthermore, splanchnic artery occlusion shock produced cardiovascular changes, reduced macrophage phagocytosis (23 +/- 4.6%) and killing (6 +/- 1.1%) activities, and induced a massive necrosis of the ileum. A passive immunization with a hyperimmune serum containing antibodies against murine TNF-alpha significantly protected rats from the lethal effects of splanchnic artery occlusion shock, lowered serum TNF-alpha (6 +/- 2.1 U/ml), and completely reverted the impairment in peritoneal macrophage phagocytosis (48 +/- 4.8%) and killing (13 +/- 1.5%) activities. In addition passive immunization had beneficial effects on the cardiovascular changes occurring during splanchnic artery occlusion shock and prevented necrosis of the ileum induced by this model of shock. By contrast, pretreatment with polymyxin B, an "antiendotoxin" antibiotic, did not modify the lethal effects and the TNF-alpha production induced by splanchnic artery occlusion shock. Furthermore, endotoxin was undetectable in the blood of splanchnic artery occlusion shocked rats. These findings are consistent with the involvement of TNF-alpha in splanchnic artery occlusion shock and suggest that the cytokine represents an important mediator of non-septic shock.