BACKGROUND Continuous intravenous administration of adenosine to humans often results in a paradoxical rise in pulmonary capillary wedge pressure (PCWP), whereas arterial resistance is lowered and cardiac output and heart rate increase. This is believed to be due to diastolic stiffening of the ventricle or to a negative inotropic effect. In the present study, we tested these and other mechanisms by using pressure-volume (PV) analysis and echocardiography. RESULTS Fifteen patients with normal rest left ventricular function underwent cardiac catheterization and received adenosine at a rate of 140 micrograms/kg per minute IV for 6 to 10 minutes. PV relations were measured in 9 patients (without coronary artery disease) using the conductance catheter method. In 6 additional patients with coronary artery disease, echocardiograms were used to assess wall thickness and function, and aortic and coronary sinus blood, lactate, oxygen, and adenosine levels were measured. Adenosine increased PCWP by 19% (+2.6 mm Hg) in both patient groups while lowering arterial load by 30% and increasing cardiac output by 45% (all P < .001). There was no significant effect of adenosine on mean linear chamber compliance or monoexponential elastic stiffness, as the diastolic PV relation was unchanged in most patients. Diastolic wall thickness also was unaltered. Thus, the PCWP rise did not appear to be due to diastolic stiffening. Adenosine induced a rightward shift of the end-systolic PV relation (ESPVR) (+12.7 +/- 3.7 mL) without a slope change. This shift likely reflected effects of afterload reduction, as other indexes (stroke work-end-diastolic volume relation and dP/dtmax at matched preload) were either unchanged or increased. Furthermore, this modest shift in ESPVR was more than compensated for by vasodilation and tachycardia, so reduced systolic function could not explain the increase in PCWP. There also was no net lactate production to suggest ischemia. Rather than arising from direct myocardial effects, PCWP elevation was most easily explained by a change in vascular loading, as both left ventricular end-diastolic volume and right atrial pressure increased (P < .05). This suggests that adenosine induced a redistribution of blood volume toward the central thorax. CONCLUSIONS PCWP elevation in response to adenosine primarily results from changes in vascular loading rather than from direct effects on cardiac diastolic or systolic function.