In 36 hemodynamically stable septic patients, we explored whether changes in gastric mucosal-arterial PCO(2) gradient (PCO(2)gap) induced by a short-term dobutamine infusion may reveal hepatosplanchnic hypoperfusion. Hepatosplanchnic blood flow (HSBF) was determined by the continuous indocyanine green infusion technique and gastric mucosal PCO(2) (Pg(CO(2))) by saline tonometry. In each patient, hemodynamic measurements, blood samples, and Pg(CO(2)) determinations were performed three times: first at baseline (DOB 0), second during a dobutamine infusion at a dose of 5 microgram/kg/min (DOB 5), and third at a dose of 10 microgram/kg/min (DOB 10). The results were analyzed by Wilcoxon's matched-pairs signed rank test and are presented as medians with ranges. The PCO(2)gap decreased preferentially in groups of patients with inadequate hepatosplanchnic perfusion, i.e., with a low fractional HSBF (HSBF/CI), defined as the ratio of the HSBF to the simultaneous cardiac index, or a high gradient between the mixed venous blood and the suprahepatic blood O(2) saturations (DSvh(O(2))). In the 11 patients with a DSvh(O(2)) above 20% at baseline, PCO(2)gap decreased from 12.1 (6.3 to 19.5) mm Hg at DOB 0 to 6.2 (2.5 to 19. 3) mm Hg at DOB 5 (p < 0.001 versus DOB 0), and to 4.2 (0.1 to 35.9) mm Hg at DOB 10 (p < 0.05 versus DOB 5), whereas in the 25 patients with a DSvh(O(2)) below 20% at baseline, PCO(2)gap did not change significantly. At no time was the PCO(2)gap correlated with HSBF/CI or DSvh(O(2)). We conclude that although the PCO(2)gap does not correlate well with global indexes of gut oxygenation, such a simple dobutamine infusion test could identify patients with inadequate hepatosplanchnic perfusion.