Background. We tested the hypothesis that some of the maintenance of resting, regional hepatosplanchnic perfusion that is mediated by endogenous adenosine (ADO) during sepsis is interdependent with nitric oxide (NO). Materials and methods. Twenty-four hours after sepsis/sham induction, rats were divided into two groups. Group 1 received a 10-min infusion of the ADO antagonist 8-sulfophenyltheophylline (8-SPT; 0.9 mg/ kg . min), followed by 10 min of 8-SPT plus L-NAME (0.5 mg/kg . min). Group 2 received L-NAME first followed by 8-SPT in the presence of L-NAME (all groups: n = 6-10). Hemodynamic and regional hepato-splanchnic blood flow measurements were made prior to infusions, 10 min after initiation of each single agent infusion, and again after double agent infusion. Results. Twenty-four hours after sepsis hepatosplanchnic blood flow was significantly elevated, compared to nonseptic rats. Both ADO receptor blockade alone and NOS inhibition alone decreased total hepato-splanchnic blood flow to nonseptic values. Decreases in small intestinal and cecal blood flow accounted for the majority of this decrease, but decreased hepatic arterial per fusion contributed as well. No further alterations were seen when 8-SPT was infused in the presence of L-NAME, nor when L-NAME was infused in the presence of 8-SPT. Conclusions. These data indicate that there is significant interdependence of endogenous NO and ADO in maintaining resting small bowel, cecal, and hepatic arterial perfusion during sepsis. The lack of responses in other regions of the hepato-splanchnic circulation demonstrate regional specificity of this ADO-NO inter-dependence. (C) 2000 Academic Press.