Objective: To investigate the association between levels of nitric oxide metabolites and the cardiovascular response to endotoxin shock; and to compare the effects of dopamine, noradrenaline and S-methylisothiourea sulphate (SMT), a relatively selective inducible nitric oxide synthase inhibitor, on haemodynamics, oxygen transport, markers of global and regional perfusion and survival in an immature pig model of endotoxin shock. Design: Randomised, blinded laboratory study. Setting: Animal research laboratory in a university hospital. Subjects: Three- to four-week-old pigs. There were eight animals in each of four groups. Interventions: The animals were anaesthetised and ventilated. Peripheral arterial, pulmonary artery and central venous cannulae and a gastric tonometer were inserted. Endotoxin shock was induced by a continuous infusion of E. coli lipopolysaccharide. Animals were randomly allocated to one of four groups: i) dopamine at 5-20 μg/kg/min, ii) noradrenaline at 0.1-0.5 μg/kg/min, iii) SMT at 20-80 μg/kg/min, or iv) placebo, to which the investigators were blinded. The study drug was commenced 1 hour after initiation of endotoxaemia, and then infused continuously for 6 hours. Volume resuscitation was given to all groups throughout the study to maintain a pulmonary artery wedge pressure at 2-4 mm Hg above pre-endotoxin baseline levels. Measurements: At baseline after cannulation, 1 hour after the initiation of endotoxaemia, and every 90 minutes after the commencement of the study drug infusion, the following data were recorded: heart rate, systemic blood pressure, mean pulmonary artery pressure (PAP) and pulmonary artery wedge pressure (PAWP), central venous pressure (CVP), gastric intramucosal pH (pHi), base deficit, blood lactate, mixed venous oxygen saturation (SvO2), cardiac index, systemic and pulmonary vascular resistance indices (SVRI and PVRI) and oxygen delivery (DO2). Blood was collected for assay of combined serum nitrate and nitrite (nitrogen oxides). Endpoints: The major endpoints were: the relationship between background nitric oxide (NO) activity and initial haemodynamic response to endotoxin; and the relative effect of the study drugs over time on: DO2, SVR, PVR, blood lactate, pHi, SvO2 and base deficit. Results: There was a significant, albeit weak, relationship between baseline serum nitrogen oxide concentrations and the initial cardiovascular response to endotoxin: levels of serum nitrogen oxides were positively associated with change in cardiac index; and negatively associated with the change in SVRI and PVRI. SMT produced greater increases in SVRI than the other three treatment arms, but caused a decrease in DO2. Dopamine increased DO2 in the early stages, but not later in the course of the study. SMT produced a greater increase in PVR than dopamine. Conclusions: Endogenous NO activity may be one determinant of the cardiovascular response to infection. Use of SMT in endotoxin shock resulted in vasoconstriction and reduced DO2, compared with catecholamines or volume alone.