Effects of Milrinone on Rest and Exercise Hemodynamics in Heart Failure With Preserved Ejection Fraction

Heart failure with preserved ejection fraction (HFPEF) accounts for a substantial proportion of the population of patients with heart failure, and the prevalence is rising (1). The pathophysiology of HFPEF is complex (2); a rapid rise in left ventricular diastolic and pulmonary pressures particularly during exertion is a fundamental feature (3). Interventions with a broad range of pharmacological agents, including those targeting the nitric oxide and cyclic guanosine monophosphate pathways, have largely been neutral in HFPEF (1,4). As such, identifying effective treatments for patients with HFPEF remains a major therapeutic challenge.

We investigated the hypothesis that milrinone, a type III phosphodiesterase inhibitor, would exert a favorable effect on hemodynamics in patients with HFPEF. This concept was based on the theoretically favorable actions of milrinone on key pathophysiological targets, including left ventricular diastolic performance, pulmonary pressures, and potentially systolic function.

Patients with HFPEF (n = 20; age: 68 ± 2 years; left ventricular ejection fraction 64 ± 2%) underwent right heart catheterization and radial arterial cannulation for measurement of hemodynamics at rest and during peak symptom-limited exercise, as described previously (3). After the baseline evaluation, subjects were randomly allocated, in a blinded manner, to receive either milrinone (50 μg/kg intravenously over 20 min) or saline. At the conclusion of the infusion, rest and exercise measures were repeated. Exercise was conducted at individual matched workloads and durations. The primary objective was to compare the between-group changes in exercise hemodynamics after administration of placebo versus milrinone. The study was approved by the Alfred Hospital Research and Ethics Committee, and participants provided written informed consent. Data are given as mean ± SEM.

Before infusions, there were no differences in resting hemodynamics. Pretreatment exercise hemodynamic responses (presented as delta values for pre-placebo vs. pre-milrinone) were generally not significantly different: heart rate (28 ± 5 vs. 30 ± 3 beats/min), mean arterial pressure (21 ± 4 vs. 22 ± 3 mm Hg), mean pulmonary artery pressure (17 ± 2 vs. 23 ± 2 mm Hg), cardiac index (1.8 ± 0.2 vs. 2.3 ± 0.2 l/min/m²), and stroke volume index (6 ± 1 vs. 4 ± 1 ml/m²). The pulmonary capillary wedge pressure rise with exercise (PCWP) was greater in the pre-milrinone group (15 ± 1 vs. 20 ± 5 mm Hg; p < 0.05). Infusion of vehicle was without any effect on resting or exercise hemodynamics. Intravenous milrinone caused significant reductions in resting right atrial pressure (7 ± 1 vs. 3 ± 1 mm Hg; p = 0.001), mean pulmonary artery pressure (23 ± 2 vs. 19 ± 3 mm Hg; p < 0.01), and PCWP (13 ± 1 vs. 8 ± 1 mm Hg; p < 0.001). Resting heart rate (70 ± 3 vs. 77 ± 4 beats/min; p = 0.001) and cardiac index (2.8 ± 0.1 vs. 3.1 ± 0.1 mm Hg; p < 0.05) increased, and mean arterial pressure was unchanged. As shown in Figure 1, milrinone substantially attenuated the rise in right atrial pressure, mean pulmonary artery pressure, and PCWP during exercise compared with vehicle, while the peak exercise heart rate was higher. Stroke volume index at rest or during exertion was unaltered by milrinone compared with respective baseline values.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Effect of Milrinone on Exercise Hemodynamics in HFPEF

Bar graphs represent the change (vs. baseline) in peak exercise right atrial (RA) pressure, mean arterial blood pressure (BP), mean pulmonary artery (PA) pressure, pulmonary capillary wedge pressure (PCWP), cardiac index, and heart rate after vehicle or milrinone infusions. ***p < 0.001, **p < 0.01, *p < 0.05. HFPEF = heart failure with preserved ejection fraction.

We have shown that intravenous milrinone significantly reduced peak PCWP and pulmonary pressures during exercise in patients with HFPEF. To a large extent, the clinical presentation of HFPEF, principally exertional dyspnea, has been ascribed to a rapid rise in ventricular filling pressures and pulmonary pressures, most notably during exertion or fluid overload. Our study was not designed to test the effect of milrinone on exercise capacity or perceived exertion, nor was it designed to investigate longer term effects. A study of intravenous nitrite showed similar effects on exercise PCWP (5). The present findings may be explained either by a favorable effect of milrinone on ventricular distensibility or by preload reduction. Although we did not evaluate left ventricular mechanics in detail (e.g., using conductance catheters), the present findings of a fall in filling pressures with no change in stroke volume would be consistent with improved left ventricular compliance.

The present study suggests that phosphodiesterase type III inhibitors such as milrinone might be a therapeutic option for patients with HFPEF; however, randomized controlled trials are required. We observed a small increase in heart rate, the implications of which are uncertain; only 25% of our cohort were treated with beta-blockers. Although milrinone is known to exhibit pro-arrhythmic effects in patients with HFREF, the potential for such an effect in HFPEF is unknown. Taken together, further studies on the utility of milrinone in patients with HFPEF are warranted, including the role of adjunctive therapies such as beta-blockers.

• American College of Cardiology Foundation