A-HeFT

I have to take you back in history a little bit. In 1973, we wrote an article in Circulation in which we hypothesized a role for the peripheral circulation in controlling output from the heart; we called this “impedance affecting left ventricular (LV) performance.” That led us to begin using vasodilator drugs to treat severe heart failure. Another paper, in the New England Journal of Medicine in 1974, reported the dramatic hemodynamic response to nitroprusside infusion in refractory heart failure. We thought we were onto something; perhaps vasoconstriction was a primary factor in the progression of heart failure.

As we looked at oral therapies that would do the same thing as nitroprusside, we investigated a combination of hydralazine and isosorbide dinitrate; when we administer these 2 drugs together, the effect is exactly comparable to an infusion of sodium nitroprusside. And we moved on to do a clinical trial to see whether this oral therapy would alter the long-term course of heart failure: the V-HeFT (Vasodilator Heart Failure Trial). There were only 600-and-some patients in the whole trial; everybody was on digoxin and diuretics, and they were randomized to 1 of 3 arms: 1 group got placebo, the 2nd group got a combination of isosorbide dinitrate and hydralazine, and the 3rd group got the α-blocker prazosin, also a vasodilator. In V-HeFT, there appeared to be a preferential response to the combination of nitrate and hydralazine over vasodilation with prazosin alone, with a significant reduction of cumulative mortality.

We also followed sequential radionuclide angiocardiography in these patients, looking at the ejection fraction over time; over 4 years, the patients on placebo and the patients on prazosin had a progressive decline of ejection fraction, while the patients taking nitrates and hydralazine showed an improvement in ejection fraction. This occurred despite the fact that prazosin effectively lowered blood pressure more, over time, than did the nitrate–hydralazine combination. It was not just the blood pressure that affected LV function. As we looked closer, we came to appreciate that a lot of what was going on was LV remodeling. In the placebo and prazosin patients, there was an increase over time in LV size that was associated with a reduced ejection fraction. The nitrate–hydralazine combination somehow favorably affected progressive structural change in the left ventricle.

Neurohormonal Balance

We know that hormones are activated in heart failure. Most of these, like norepinephrine, renin, vasopressin, atrial peptides, endothelin-1, are vasoconstrictors, except for the peptides; and they all are growth stimulators. We thought that hormonal effects were major contributors to the progressive remodeling of the left ventricle; that led to years of study of neurohormonal inhibitors—ACE inhibitors, angiotensin receptor blockers, β-blockers, aldosterone inhibitors—all aimed at inhibiting the remodeling process that was dependent on this neurohormonal activation.

What we came to recognize was that there are really 2 factors affecting the output and function of the left ventricle; one is functional, and one is structural. The functional effect, which is an increase of impedance, can be improved by any vasodilator and will relieve symptoms, as it did with nitroprusside and as it does with a combination of nitrate and hydralazine. But there are also structural changes that take place that we were not aware of when we began the V-HeFT series of studies. This structural effect appears to also be dependent on hormonal stimulation, and it relates to remodeling of the left ventricle, which reduces the ejection fraction and leads to progression of the disease and shortened life expectancy.

What is so special about nitrates and hydralazine? When we began these studies, we had no clear evidence of how this combination worked, except as a vasodilator; we have come to realize that nitric oxide probably played a key role in the process. Isosorbide dinitrate is a vasodilator of both the large and small arteries and the veins, but it is also a nitric oxide donor. Hydralazine is an arterial dilator, and has been used in hypertension for many years, but it is also an antioxidant and inhibits the breakdown of nitric oxide. When you give the 2 together, you get the equivalent of nitric oxide enhancement.

In addition to ventricular remodeling, there is remodeling of the vasculature over time; this remodeling process also appears to be related to nitric oxide. When there is impaired nitric oxide release, you get platelet aggregation, a decrease in vascular compliance, and an increase in vascular tone. On the other hand, in the presence of nitric oxide depletion, you get hypertrophy and hyperplasia of vascular smooth muscle, and, ultimately, atherosclerosis. Nitric oxide is a powerful inhibitor of vascular changes, as well as cardiac changes. The progressive remodeling process in the myocardium and the vasculature is stimulated by neurohormonal activation, frequently by angiotensin, which, interestingly enough, is inhibited by nitric oxide. So what we really need in order to maintain our cardiovascular health is a balance between angiotensin and nitric oxide.

Unique Features of African-Americans

There were close linkages between nitric oxide and endothelial dysfunction, and there was growing evidence that African-Americans have more endothelial dysfunction and a less active nitric oxide system than do non–African-Americans. We went back to the original V-HeFT data and identified 180 patients who claimed to be African-American and compared them with the patients who claimed not to be African-American. In the African-American patients, there was a striking reduction in cumulative mortality with a combination therapy of nitrate and hydralazine, and there was no demonstrable difference in the patients who were not African-American.

We looked retrospectively at the SOLVD database and noted that the response to ACE inhibitors appeared to be less dramatic in blacks. There were 800 African-Americans in SOLVD; we matched them as well as we could, on the basis of known risk factors, with a comparable group of white patients. Mortality and the need for hospitalization were both higher in the blacks. There was no difference in mortality between blacks and non-blacks in the response to enalapril versus placebo, but there was a striking reduction in the hospitalization rate in non-black patients taking enalapril, and no benefit in hospitalization in black patients taking enalapril. This has also been shown to be true in β-blocker studies; these drugs don't appear to work as well in blacks.

The A-HeFT Study

This set the stage for the A-HeFT (African-American Heart Failure Trial), which involved African Americans with a low ejection fraction and a dilated left ventricle. Ninety-seven percent of these patients were in New York Heart Association (NYHA) class III, their blood pressures were within normal limits, and the average ejection fraction was 24%. These were well treated individuals; 70% were on an ACE inhibitor; 17% were on an angiotensin-receptor blocker; almost three quarters of them were on a β-blocker; a fair percentage were on spironolactone; and most of them were on a diuretic. We randomized these patients, about 500 in each group, to BiDil (a fixed-dose combination of isosorbide dinitrate and hydralazine, the same one used in V-HeFT) or placebo. The primary endpoint was a complex symptom-scoring system. But what was really striking were the mortality outcomes in these 2 populations. There was a 43% reduction in mortality in the BiDil group, compared with the placebo group. It was so dramatic that the Data Safety and Monitoring Board terminated the trial prematurely because of the efficacy of the drug combination; they did not feel that it was ethical to continue the placebo arm of the study. The primary endpoint was strikingly positive in favor of isosorbide dinitrate and hydralazine, and there was a striking reduction in the need for heart-failure hospitalization, together with significant improvements in quality of life. In regard to side effects, severe headache occurred in about 5% of the patients on BiDil, compared to 1% on placebo. There was a little more dizziness with BiDil. Exacerbations of heart failure were significantly more common in the placebo group.

Putting It Together

So where are we today? I believe our old idea that effective therapy for heart failure should include blockade of the renin-angiotensin-aldosterone system and β-adrenergic blockade now has to be supplemented by the concept of nitric-oxide–enhancing therapy with fixed-dose isosorbide dinitrate and hydralazine. In mechanical terms, if you're normal, you have a balance between growth/remodeling and suppression of that growth/remodeling. When you develop heart failure, there is neurohormonal stimulation, which, in turn, requires neurohormonal inhibitors to balance it. There may also be endothelial dysfunction and nitric oxide deficiency, a condition that requires nitric oxide enhancers to balance it. However, everyone is not the same. We know that in blacks, the only population studied in A-HeFT, there appears to be a greater role for nitric oxide deficiency and greater benefits with nitric-oxide–enhancing therapy than in non-blacks. That probably will be true in other populations but has not yet been studied prospectively. At present, we are searching for genetic markers that can tell us more about how to identify the responders: who they are, and why they respond. At this point, all we have is self-proclaimed racial background; we hope to add to that in the very near future.

Summary

In conclusion, reduced nitric oxide bioactivity contributes to cardiac and vascular remodeling. Nitric-oxide–enhancing therapy can slow cardiac and vascular remodeling. African-Americans with heart failure exhibit nitric oxide deficiency and a dramatic benefit from nitric-oxide–enhancing therapy. The potential benefit of this therapy in other groups with heart failure and in other vascular remodeling disease processes requires further study.