Inotropic and vasodilating properties of amrinone depend on the mode of administration

Intravenous infusion of amrinone at increasing rates (10-100 micrograms kg-1 min-1, N = 8) caused dose dependent decreases in left ventricular filling pressure (up to 22%, P less than 0.05) and cardiac output (up to 16%, P less than 0.05), but had no effect on heart rate, max LVdP/dt or total systemic vascular resistance in anaesthetized open-chest pigs. Despite unchanged total systemic vascular resistance, tissue vascular resistance decreased significantly in the heart, stomach, adrenals and kidneys. Although transmural left ventricular perfusion was not influenced by amrinone, coronary blood flow was redistributed in favour of the epicardium, as endo-epi blood flow ratio decreased from 0.95 +/- 0.03 to 0.82 +/- 0.03 (P less than 0.05). In the same model an intravenous bolus of 1 mg kg-1, followed by a continuous infusion of 50 micrograms kg-1 min-1 (N = 8), caused immediate changes (P less than 0.05) in left ventricular filling pressure (-35%), max LVdP/dt (+55%), heart rate (+15%) and total systemic vascular resistance (-15%). The changes in filling pressure and systemic vascular resistance persisted during the next 25 minutes, but maxLVdP/dt returned gradually to baseline in spite of increasing plasma concentrations of the drug. In the conscious pig (N = 4), administration of a 1 mg kg-1 bolus in the pulmonary artery led to similar increases in heart rate (15%) and max LVdP/dt (26%), while systolic left ventricular pressure was not affected. Direct infusion into the left anterior descending coronary artery (10-40 micrograms kg-1 min-1, N = 5) had negligible effects on overall haemodynamics and regional myocardial function. The only significant changes were a vasodilation in the coronary vascular bed accompanied by dose dependent increases in the coronary venous O2-content. From our study it appears that a bolus injection, as given in the clinical setting, is required to elicit an increase in max LVdP/dt and arterial vasodilation but that the effect on left ventricular preload is not sensitive to different modes of administration.