Mechanisms involved in the response to prolonged infusion of atrial natriuretic factor in patients with chronic heart failure.
We examined the mechanisms involved in the cardiovascular and renal response to prolonged infusion of atrial natriuretic factor (ANF) in patients with chronic heart failure. ANF infusion was titrated to produce a 30% decrease in pulmonary capillary wedge pressure or a 20% increase in cardiac output, and this dose (average, 75 +/- 4 ng/kg/min) was then administered for 20 hours. The short-term response to ANF included significant reductions in central filling pressures, increases in cardiac output, modest increases in diuresis and glomerular filtration rates, significant reduction in plasma aldosterone levels, and a 3.6-fold increase in plasma cyclic GMP levels. During prolonged infusion, plasma cGMP levels and cardiac output gradually returned to baseline. Similarly, the initially increased diuretic effects were completely abolished during prolonged ANF infusion, although plasma alpha-hANF levels remained consistently elevated above baseline values (control, 198 +/- 38; titration, 2,760 +/- 596; 20 hours, 3,499 +/- 659 pg/ml). Four hours after beginning the ANF infusion, marked increases in hematocrit levels were noted (42.5 +/- 1.0% versus 45.3 +/- 1.4%, control and infusion, respectively, p less than 0.05); during this time, no change in total plasma protein concentration occurred, indicating extravascular shift of fluid and plasma proteins. No evidence was noted for activation of vasoconstrictor hormones during prolonged ANF infusion, although mean arterial pressure was significantly reduced throughout the infusion period. Plasma pro-ANF (31-67) levels, determined as a marker for endogenous ANF secretion, were significantly suppressed as were the reductions of central filling pressures. After ANF discontinuation, heart rate and pulmonary capillary wedge pressure increased significantly above baseline values without evidence for sympathetic stimulation. We conclude that 1) prolonged infusion of ANF causes only transient increases in plasma cGMP levels but a sustained reduction of the cardiac release of ANF and that 2) the beneficial hemodynamic effects of ANF, that is, unloading of the ventricles, may be associated with or, in part, may be secondary to a shift of plasma constituents into the extravascular space. The latter may limit the therapeutic potential of ANF for long-term treatment.