Supplemental L-Arginine in Patients with heart Failure from aha journals

Many studies have demonstrated that patients with heart failure have reduced peripheral blood flow at rest and during exercise. Several studies have also demonstrated subnormal peripheral vasodilatation in response to endothelium-dependent vasodilators in heart failure. These abnormalities have been related to the severity of the symptoms patients experience and to exercise capacity. Furthermore, improvements in exercise performance with a converting enzyme inhibitor have been associated with increases in peripheral blood flow during exercise. Interventions that augment endogenous nitric oxide activity over the long term may also improve peripheral blood flow and have clinical benefits in patients with heart failure.

Endothelial cells produce nitric oxide from l-arginine via nitric oxide synthase. Studies using tracer amounts of [¹⁵N]l-arginine demonstrated that ≈1% of dietary l-arginine appears in the urine as labeled nitrate, but the effect of changes in dietary l-arginine on vascular nitric oxide formation is not known. Under normal conditions, an excess of l-arginine is available for endothelial cells, on the basis of in vitro determinations of nitric oxide synthase kinetics. However, in vitro formation of nitric oxide in response to endothelium-dependent vasodilators can be improved by l-arginine when l-arginine stores are depleted or in the presence of l-glutamine. Furthermore, studies of patients with abnormal endothelium-dependent vasodilatation related to atherosclerosis or heart transplantation have observed significant improvements during short-term intravenous infusions of l-arginine. Intra-arterial l-arginine also enhanced the forearm vasodilator response to acetylcholine, an endothelium-dependent vasodilator, in patients with heart failure. These studies suggest that, under some conditions, supplemental l-arginine may augment nitric oxide production.

The objective of this investigation was to determine whether supplemental oral l-arginine improves peripheral blood flow in patients with heart failure at rest or when stimulated by exercise, postischemic hyperemia, or an endothelium-dependent vasodilator. In addition, measures of submaximal exercise tolerance and quality of life were examined to determine whether supplemental oral l-arginine might produce clinical benefits.

Methods

Study Design

Seventeen patients who were referred for tertiary treatment of heart failure were enrolled in a two-period randomized crossover study. All patients gave written informed consent. Before data analysis, two subjects were excluded because of protocol noncompliance manifested by excessive dietary salt intake that resulted in an acute weight gain of more than 5 pounds associated with symptomatic deterioration that required adjustments in concurrent therapy. These confounding events occurred during the l-arginine treatment periods and may have biased our estimates of the true l-arginine effects. No similar events were seen during the placebo period. Characteristics of the remaining 15 patients are described in Table 1. Diabetes mellitus, hypertension, and overt atherosclerotic peripheral vascular disease were exclusion criteria, and all participants had total cholesterol levels ≤6.2 mmol/L (240 mg/dL).

Seven patients were randomly assigned to placebo followed by l-arginine, and 8 patients were assigned to the opposite sequence of treatments. During the 6-week treatment period, patients were given either 2.8 g l-arginine hydrochloride (Tyson and Associates, Inc) twice daily (n=9) or 4.2 g three times daily (n=6). The initial dose of oral l-arginine was selected to double the estimated average daily dietary intake of l-arginine. However, we did not evaluate the actual dietary intake of l-arginine in these patients. In an attempt to detect a dose effect, we arbitrarily increased the amount of supplemental l-arginine 2.5 times after 10 subjects were enrolled. Significant effects were seen after 10 weeks of oral l-arginine treatment in animals. Six-week treatment periods were used in this study to increase the likelihood that these patients with severe heart failure would remain stable for the entire study.

Matching placebo capsules were administered during the 6-week control period to mask the identity of study capsules.

Study Assessments

The following sequence of measurements was made at the end of each 6-week treatment period. Medications were withheld on the morning of these evaluations. First, study nurses read standard instructions and then asked the patients to complete the Living With Heart Failure questionnaire to ascertain how their heart failure had affected their quality of life.The total distance covered during a 6-minute walk test was used as a measure of exercise tolerance during daily activities. A standard set of instructions was read each time the walk test was done. Patients were told to walk as far as they could in 6 minutes in an isolated 60-foot hallway. During the walk test, patients were periodically encouraged to go as far as they could. Rest stops were allowed if symptoms became too bothersome.

Vascular function was examined at the end of each 6-week period by a regionally perfused forearm technique. Forearm blood flow was measured by venous occlusion plethysmography. Patients rested in the supine position with their nondominant forearm abducted ≈30° and raised to midchest position. A mercury-in-Silastic strain gauge was placed around the upper forearm and connected to a calibrated plethysmograph to measure forearm blood flow as the rate of change of forearm volume (mL·min⁻¹·100 mL⁻¹ forearm volume). Venous occlusion was induced by rapid inflation of a blood pressure cuff placed around the upper arm to 40 mm Hg (D.E. Hokanson, Inc). Hand blood flow was excluded by a wrist cuff inflated to suprasystolic pressure before each measurement, which was the average of three recordings. Room temperature was constant during each day of study and did not vary by more than 1°C between study days. Forearm volume was calculated as the volume of a truncated cone based on wrist and upper forearm circumference and forearm length. The average forearm volume was 1097±169 mL.

After measurement of baseline forearm blood flow, forearm exercise was performed with a hand dynamometer set at 15%, 30%, and 45% of each patient’s maximal voluntary contraction as has been previously described. The exercise protocol consisted of cycles of 5 seconds of contraction followed by 10 seconds of relaxation. Patients exercised for 2 minutes, then forearm blood flow was measured 5 seconds into each relaxation phase during the third minute of exercise. When flow had returned to baseline, a 5-minute period of forearm ischemia was created by inflation of the cuff around the upper arm to suprasystolic pressure. Flow measurements taken immediately after the pressure occlusion was deflated and after 15, 30, 45, and 60 seconds were used to characterize the amount of reactive hyperemia.

A 20-gauge cannula was then inserted under local lidocaine anesthesia into the brachial artery of the same arm to study responses to intra-arterial administration of an endothelium-dependent vasodilator, methacholine, and an endothelium-independent vasodilator, nitroprusside. Patients rested quietly for 30 minutes after catheter insertion before baseline blood samples, blood pressure, and heart rate were obtained. Baseline forearm blood flow was measured, followed by infusion of methacholine at 0.3, 1.5, and 3.0 μg/min, with measurement of flow at the end of each 90-second infusion. When baseline flow was reestablished, the responses to three doses of nitroprusside, 1, 5, and 10 μg/min, were determined in the same manner. Finally, a competitive inhibitor of nitric oxide synthase, l-N-monomethylarginine, was administered to assess basal nitric oxide activity. First, baseline flow was reestablished, then flow was measured at the end of two cumulative 4-minute infusions of l-N-monomethylarginine (4 and 8 μmol/min).

Vascular smooth muscle relaxation can increase arterial compliance as well as lower arteriolar resistance.Arterial compliance was estimated by pulse contour analysis.Pressure waves were obtained by placing a tonometer sensor array (model N-500, Nellcor) over the radial artery of the nondominant arm. Tonometer dispersions were calibrated by use of synchronized automatic blood pressure cuff measurements taken from the contralateral arm. Pressure waveforms from several stable beats captured over 30 seconds were fit by nonlinear least-squares regression to a modified Windkessel model resulting in two separate compliance indexes based on the exponential decay of the diastolic portion of the wave form (C₁) and the superimposed decaying sinusoidal waveform (C₂).

The effects of l-arginine on several vasoconstrictor pathways that are often activated in patients with heart failure were also examined because of previous reports suggesting that increased nitric oxide activity may interact with the sympathetic nervous system or endothelin. Plasma endothelin was quantified by radioimmunoassay.The rabbit antibody used (Peninsula Laboratories, Inc) had 100% cross-reactivity with endothelin-1, 7% with endothelin-2 and -3, and 17% with big endothelin. The normal range for endothelin in our laboratory is 0.2 to 1.9 pmol/L. Plasma norepinephrine was measured by high-performance liquid chromatography with electrochemical detection.Our normal plasma norepinephrine ranges from 0.8 to 1.9 nmol/L. Plasma renin activity was determined by a radioimmunoassay with a normal range from 0.12 to 0.85 nmol·h⁻¹·L⁻¹. The concentration of l-arginine in plasma was determined by our hospital laboratory, in which normal values range from 20 to 180 μmol/L.

Safety measures included clinic visits every 3 weeks; blood tests for electrolytes, glucose, insulin, renal function, liver function; and complete blood counts. Intravenous administration of much higher doses of l-arginine hydrochloride have been associated with hyperinsulinemia and changes in potassium concentrations.If similar changes occur during oral l-arginine treatment, endothelium-dependent vasodilatation could be affected. Blood urea nitrogen was also increased by oral l-arginine in a previous animal study.

Statistical Analysis

Data were summarized as mean±SD except in the figures, in which standard error bars are shown. Measurements obtained after 6 weeks of oral l-arginine were compared with those obtained after the 6-week placebo control period by ANOVA for repeated measurements that included estimates of the effects of treatment sequence and dose of oral l-arginine. Since treatment sequence and l-arginine dose did not significantly affect the results, the data were not presented for subgroups defined by these factors. If changes in neurohormones followed a skewed distribution, a Wilcoxon signed rank test was used to compare concentrations during the oral l-arginine and placebo treatment periods. A value of P≤.05 was considered statistically significant.

Results

Clinical Assessments

Supplemental oral l-arginine improved patients’ functional status. There was a significant improvement, from 55±28 to 42±26 (P<.05), in the Living With Heart Failure questionnaire score (Fig 1) and a significant increase, from 390±91 to 422±86 m (P<.05), in the distance walked in 6 minutes (Fig 2). Body weight remained stable (87.7±11.4 versus 87.2±11.4 kg) throughout the 12-week study period. Mean supine blood pressure decreased slightly, from 82±9 to 80±6 mm Hg (P<.05), during l-arginine administration. Supine resting heart rate did not change (81±16 versus 81±13 bpm).

Peripheral Vascular Function

During forearm exercise, forearm blood flow increased significantly after the l-arginine treatment period compared with the placebo period (Fig 3). The average improvement in forearm blood flow during exercise was 1.5±3.0 mL·min⁻¹·dL⁻¹ (P<.05). Maximum voluntary hand contraction did not change (35±9 versus 36±10 kg). In contrast to exercise hyperemia, reactive hyperemia after 5 minutes of arterial occlusion was not significantly augmented (Fig 4).

Methacholine was used to examine the effect of oral l-arginine on the response of the endothelial nitric oxide pathway when stimulated to release increased amounts of nitric oxide. Increases in forearm blood flow in response to methacholine were on average 1.5±4.2 mL·min⁻¹·dL⁻¹ greater with l-arginine than with placebo (Fig 5), but this difference did not attain statistical significance (P=.33). Responses to exogenous nitric oxide provided by infusion of nitroprusside were similar during the l-arginine and placebo periods (Fig 6).

Vasoconstrictor responses to l-N-monomethylarginine, an inhibitor of nitric oxide synthase, were similar during the l-arginine and placebo treatment periods (Fig 7), suggesting that basal endothelial nitric oxide activity was not increased by oral l-arginine supplementation. This finding is consistent with unchanged forearm blood flow under basal conditions during l-arginine administration compared with placebo (3.7±1.1 versus 3.8±1.6 mL·min⁻¹ · dL⁻¹, respectively).

Arterial compliance as assessed by pulse contour analysis was improved by administration of oral l-arginine. Compliance estimated from the exponential decay of the diastolic pressure (C₁) improved to 2.36±0.30 from 1.99±0.38 mL/mm Hg (P<.001), and the second compliance term in the modified Windkessel model of the circulation (C₂) increased from 0.048±0.024 to 0.064±0.026 mL/mm Hg (P<.05).

Biochemical Assessments

There was a significant increase in mean plasma l-arginine concentration, from 85±21 to 98±28 μmol/L (P<.05), during administration of supplemental oral l-arginine. Other data are summarized in Table 2. Small increases in blood urea nitrogen were seen, which may be related to l-arginine metabolism in the urea cycle. There were no differences in serum electrolytes.

Endothelin levels fell significantly, from 1.9±1.1 to 1.5±1.1 pmol/L (P<.05). Plasma norepinephrine and plasma renin activity were not significantly reduced by oral l-arginine.

Adverse Effects

One patient experienced diarrhea during the placebo period. Two patients, one in each treatment period, had acute episodes of gout during the trial. End-point assessments were avoided during these attacks. No other potential adverse effects were noted.

Discussion

This is the first randomized, double-blind study that investigated whether supplemental oral l-arginine could improve peripheral blood flow and functional status in patients with heart failure. The results suggest that oral l-arginine has beneficial effects in patients with moderate to severe heart failure. Patients noted significantly fewer limitations secondary to their heart failure as assessed by the Living With Heart Failure questionnaire and walked farther during a 6-minute walk test. In addition, limb blood flow during exercise and arterial compliance were significantly enhanced by oral l-arginine compared with placebo. Finally, circulating levels of the potent vasoconstrictor endothelin were reduced by l-arginine.

Previous studies in animals and humans with heart failure have indicated that short-term administration of l-arginine can restore vasodilator responses to an endothelial nitric oxide–dependent vasodilator.This was the first study to examine the effects of oral l-arginine administered over several weeks. Three different assessments were used to determine whether supplemental oral l-arginine could improve peripheral blood flow in patients with heart failure when nitric oxide activity was stimulated. Both forearm exercise and release of a brachial artery occlusion increase shear stress in the forearm vasculature, which can lead to increased formation of nitric oxide from l-arginine.Muscarinic agonists have served as a pharmacological stimulus to examine the functional status of the endothelial nitric oxide pathway ever since the seminal study of Furchgott and Zawadzki. Significant enhancement of exercise hyperemia was observed during l-arginine administration and may have been related to increased nitric oxide activity. However, additional investigations using inhibitors of nitric oxide synthase during exercise are needed to determine whether the observed increase in exercise hyperemia was indeed mediated via the nitric oxide pathway. The effect of supplemental oral l-arginine on methacholine-induced vasodilation was, on average, similar to the increase in exercise hyperemia, but the changes in methacholine responses were more variable than changes in exercise hyperemia. Given greater variation in the effect of l-arginine on methacholine responses, a much larger sample size would be needed to achieve statistical significance and provide a precise estimate of any effect of l-arginine. Postischemic hyperemia was not significantly increased by oral l-arginine. Although peripheral blood flow tended to improve during all three interventions used to stimulate the nitric oxide pathway, the results were somewhat inconsistent, perhaps because of differences in the extent to which nitric oxide contributes to the total vasodilator responses to these three interventions. Overall, these results can neither establish nor preclude the possibility that supplemental l-arginine may be converted to nitric oxide in patients with heart failure when activity of the nitric oxide synthase pathway is increased by different stimuli.

Lack of significant increases in the vasoconstrictor response to l-N-monomethylarginine, an inhibitor of nitric oxide synthase, suggests that oral l-arginine did not substantially enhance basal nitric oxide formation in the forearm vasculature. Resting vasoconstrictor tone as measured by forearm vascular resistance also was not reduced, since basal forearm blood flow did not increase and mean blood pressure was only slightly decreased. Similar responses to nitroprusside during the l-arginine and placebo treatment periods suggest that l-arginine did not alter nitric oxide–induced smooth muscle relaxation.

Indexes of vascular compliance were also significantly improved during the l-arginine treatment period. Reduced vascular compliance has been documented in patients with heart failure by pulse contour analysis as well as other methods.Theoretically, improvements in peripheral vascular compliance could decrease reflection of pressure waves from the periphery, thereby reducing impedance to left ventricular outflow. In addition, better arterial compliance may improve arterial baroreceptor function and decrease sympathetic activity in patients with heart failure. Other studies have suggested that endothelial nitric oxide can attenuate sympathetic nervous system activity.However, mean plasma norepinephrine concentration was not reduced by l-arginine. This observation, in conjunction with the lack of change in basal forearm vascular resistance, suggests that compliance was not improved by reductions in sympathetic vasoconstrictor tone.

Few reports were found in the literature describing structural alterations in arteries from patients with heart failure. Theoretically, increased circulating levels of growth-promoting factors such as norepinephrine, angiotensin II, and endothelin may alter vascular structure. In infarcted rats, increased medial thickness of intramuscular resistance vessels and increased carotid artery stiffness associated with increased collagen content and adventitial thickness have been demonstrated. It is possible that l-arginine could have improved vascular compliance via structural changes related to antigrowth activity of nitric oxide and barrier functions of intact endothelium.Indexes of arterial compliance based on pulse contour analysis cannot be used to isolate the contributions of hemodynamic versus structural changes. l-Arginine had only minimal effects on blood pressure that would not be expected to greatly improve vascular compliance through a decrease in distending pressure. Nitric oxide has been shown to inhibit the positive inotropic effect of β-adrenergic stimulation. Since our measures of arterial compliance vary in direct proportion to the cardiac output and cardiac output was not directly measured, it is possible that the compliance estimates were confounded by changes in cardiac output. More in-depth investigation of changes in compliance with l-arginine is needed.

Plasma endothelin levels were significantly reduced during the l-arginine treatment period compared with the placebo period. This study did not examine whether decreased formation or increased clearance of endothelin from the circulation could explain this observation. Interestingly, changes in the availability of nitric oxide have been associated with the regulation of endothelin production in vitro. Since endothelin is a potent vasoconstrictor and mitogenic factor, reduced production of endothelin may be beneficial in patients with heart failure. A recent study of an endothelin receptor antagonist administered to patients with severe heart failure observed significant hemodynamic improvements.These investigators also confirmed a strong relationship between plasma endothelin and mean pulmonary vascular resistance. More research is needed to determine whether this potential mechanism accounts for some of the effects of oral l-arginine.

Others have shown that oral l-arginine can improve responses to an endothelium-dependent vasodilator in patients and animals with hypercholesterolemia in the absence of heart failure. Since the precise degree of elevation of cholesterol or its derivatives, such as oxidized LDL cholesterol, that are needed to alter the activity of the nitric oxide pathway have not been established, it is possible that the effects of l-arginine seen in this study were related to cholesterol.However, l-arginine did not change total circulating cholesterol and had similar effects on exercise blood flow, methacholine-induced vasodilation, 6-minute walk test, and quality of life when subgroups defined by the median cholesterol level (4.7 mmol/L, or 180 mg/dL) or a history of ischemic coronary artery disease were examined (data not shown).

Oral l-arginine, 7 g three times a day for 3 days, did not enhance endothelium-dependent forearm vasodilatation in normal subjects who had relatively high cholesterol levels (mean, 5.4 mmol/L).Interestingly, platelet aggregation was inhibited by oral l-arginine. The effect of l-arginine on platelet aggregation could be blocked with l-N-monomethylarginine, suggesting that nitric oxide synthase mediated this effect. Platelet aggregation was not evaluated in our study.

Although this investigation was limited to a small sample from a single medical center, the observation of multiple improvements, including increased quality of life, submaximal exercise performance, forearm exercise hyperemia, arterial compliance, and reduced plasma endothelin, warrant further investigations of supplemental oral l-arginine in patients with heart failure. The optimal dosage regimen was not established by this study, and longer studies are needed to determine whether the l-arginine effects persist. A mechanism of action for oral l-arginine could not be established on the basis of these data, which were focused primarily on the forearm vasculature and indirect measures of the nitric oxide pathway. If therapeutic effects can be confirmed by other investigations, studies to identify the mechanism of action may provide valuable insight into the biology of l-arginine.