Abstract
Following a hypertension symposium in New York City on February 2, 2005, a roundtable was convened to discuss the use of β blockers in the management of hypertension. Dr. Marvin Moser, Clinical Professor of Medicine at the Yale University School of Medicine, New Haven, CT, moderated the discussion. Participating in this roundtable were Dr. Thomas Pickering, Professor of Medicine at the Columbia University School of Medicine, New York, NY, and Dr. William Cushman, Professor of Preventive Medicine and Medicine at the University of Tennessee College of Medicine, Memphis, TN.
DR. MOSER: Bill, there is a major debate about the use of β blockers in hypertensive patients. For many years, the Joint National Committees on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNCs) suggested diuretics or β blockers as initial therapy. The JNC I in 1977 suggested diuretics and in JNC II and III diuretics and/or β blockers were recommended. These recommendations were based on trial results. In many of these trials, however, either a diuretic or a β blocker or a combination of the two medications were given and it was difficult to determine exactly which medication was the most effective one. In one study, the Medical Research Council in the Elderly Trial (MRC), it was clear that a diuretic‐based regimen reduced coronary heart disease events more than a program based on a β blocker. Additional data from the JNCs over the years continued to suggest β blockers or diuretics as appropriate first‐step agents. In JNC VI in 1997, diuretics were recommended as a first step in the elderly with diuretics/β blockers, if needed. JNC 7 states that diuretics are preferred as initial therapy in most cases and considers β blockers as appropriate therapy in special situations.
Where do β blockers fit? Are they still one of the preferred first‐step agents? Do they lower blood pressure? Do they have a positive effect on morbidity/mortality? Are they as well tolerated as the other drugs?
DR. CUSHMAN: For many years, we only had outcome studies comparing placebo to treatment that included either diuretics or β blockers as initial therapy. It was clear, as you've mentioned, that several hypertension morbidity studies showed positive effects of the β blockers. In general, the positive studies were in somewhat younger European populations where results with these agents were similar to diuretics. It is important to mention that we actually never had a study in the United States that specifically tested β blockers in terms of outcomes in hypertension. Once we started studying and looking at older populations, the β blockers did not appear to be as effective at reducing cardiovascular events as diuretics (and several other classes), and so diuretics are clearly preferred over β blockers as initial therapy. In addition, based on recent trials, β blockers May now be less preferred as initial therapy than angiotensin receptor blockers [ARBs], calcium channel blockers [CCBs], or ACE [angiotensin‐converting enzyme] inhibitors.
The paradox about the benefits that were seen is that the use of β blockers seemed to defy pathophysiology and logic. On the one hand, these medications did lower diastolic pressure, at least in young or middle‐aged patients, seemingly as well as diuretics and other drugs. But on the other hand, here is a class of medications that, unlike virtually anything else we use for hypertension, has almost no specific beneficial effects on the arteries, and does not lower systolic blood pressure as well as many other classes of antihypertensive agents.
DR. MOSER: In fact, at least in the short term, many of the β blockers without α‐blocking capabilities tend to cause some vasoconstriction.
DR. CUSHMAN: They can vasoconstrict, but the net hemodynamic effect is usually no effect on either vasoconstriction or vasodilatation. The main effect of β blockers on blood pressure seems to be the result of reduction in cardiac output, although I believe that reduction in the generation of renin also plays a role in their blood pressure‐lowering effects.
DR. MOSER: How do β blockers reduce renin and angiotensin II? Tom, do they act the same as an ACE inhibitor or an ARB?
DR. PICKERING: Renin release is under sympathetic control. A β blocker inhibits the release of renin from the kidney, rather than blocking the effects of angiotensin II at the periphery as an ARB does, or the conversion of the nonvasoconstrictor angiotensin I to angiotensin II as an ACE inhibitor does. But, as Bill says, when you look at hemodynamic changes, certainly in the early stages of β blockade, the main effect is on cardiac output. I believe there are data suggesting that peripheral resistance comes down to some extent with chronic treatment. There is some debate as to whether or not this is from the effects of renin inhibition or whether it is due to some autoregulation effect. But one thing β blockers do is lower heart rate as they lower blood pressure. I'm certainly impressed by the reduction in cardiac output.
DR. MOSER: To sum this up—Bill is saying that for years we had good data with placebo‐controlled studies and mostly in younger or middle‐aged people, but we had very little data in the elderly. As we gathered comparative data, especially in the elderly, we found that the use of a β blocker would decrease the diastolic pressure, but not the systolic as much. Physiologically, we were concerned about using a medication that did not primarily vasodilate. All the other antihypertensive agents, including diuretics, really act as vasodilators. But, despite this, β blockers appeared to be effective, at least in some populations, in reducing cardiovascular events.
Tom, there are specific beneficial effects of β blockers, aren't there? They reduce cardiac output and work, and lower heart rate and oxygen consumption.
DR. PICKERING: Well, there is absolutely no question that β blocker use benefits people who have coronary heart disease. The mechanism for that is not too clear. It May be that they reduce sudden cardiac death. They are also obviously extremely useful and invaluable in patients with congestive heart failure. But in the hypertensive patient who does not have a major cardiac condition, their value as monotherapy is highly questionable.
There's a figure in JNC VI in 1997 that compares the β blocker trials with the diuretic‐based trials. The β blockers did not look nearly as good as the diuretic‐based treatments. Since then, new data have only reinforced that view.
DR. MOSER: Is this primarily in the elderly?
DR. PICKERING: Most of the data were in the elderly, some from Swedish trials. Part of the problem, as you mentioned, is that many of the early trials included both β blockers and diuretics in combination—where there undoubtedly was benefit.
DR. MOSER: Do you both believe that, at present, the data do not indicate that monotherapy with β blockers should be listed as one of the preferred initial approaches in patients without specific target organ problems?
DR. CUSHMAN: I think that's a controversial area. In JNC 7, there were several of us who, based on the data we had then, recommended that we not place β blockers on the same level as ACE inhibitors, CCBs, or diuretics. Certainly, diuretics should still be the preferred class. But, at the same time, we had placebo‐controlled trials with β blockers in hypertension, none with ACE inhibitors, and only one or two with CCBs.
DR. MOSER: And we're never going to have them.
DR. CUSHMAN: Right, so the argument was, we have placebo‐controlled trials, so we should include the β blockers among the preferred classes for initial therapy. But now that we have more comparative outcome studies, the data seem to suggest that the β blockers May not be quite as optimal as the other classes that we're considering for initial therapy. There also seem to be differences in what outcomes are prevented, both in individual trials and in the meta‐analyses. Interestingly, heart failure seems to benefit as much with β blockers as with other drugs.
DR. MOSER: You mean prevention of heart failure?
DR. CUSHMAN: Yes, prevention of heart failure. But it's the coronary events and strokes that do not appear to be reduced as much by β blockers as with diuretics, for example, in hypertensives. The failure to prevent coronary events is somewhat ironic. I agree with Tom. I think that if you look at the post‐myocardial infarction [MI] trials with β blockers, much of the benefit is early, within the first year or two after an MI.
DR. MOSER: Do you think that is a result of a reduction in arrhythmias?
DR. CUSHMAN: Whatever it is, it has to do with sudden death. It May be unrelated to blood pressure or vascular health. So the discussion is not that β blockers aren't good additional therapy in somebody who is at high risk for sudden death or a major cardiovascular event, but that their use May not prevent vascular disease. In the LIFE [Losartan Intervention for End Point Reduction] study, for example, the big difference between the ARB, losartan, and the β blocker, atenolol, was the failure of atenolol to prevent strokes as well as the ARB.
DR. MOSER: So what we are saying is that you can lower the blood pressure, but not as much as with other drugs. In the elderly, coronary events are not prevented as effectively as with diuretics, and in patients with LVH [left ventricular hypertrophy], strokes are not prevented as well as with an ARB. In other words, β blockers aren't shaping up to be as effective as some of the other drugs. Is that fair?
DR. PICKERING: Yes, I think so. This came up at the FDA [Food and Drug Administration] when there was an application for a special indication for an ARB based on the LIFE trial. The issue was whether the ARB was better than other drugs, or was the comparator drug (atenolol) not as good? There was a lot of debate about that because there were more strokes in the β‐blocker group. Since then, there has been a meta‐analysis of results with atenolol which suggested that even though it lowers blood pressure, it is not as good at preventing morbid events as some of the other drugs that we use. I tend not to favor starting patients on a β blocker unless there is a specific indication. I tend to use β blockers much less than I did a few years ago.
DR. MOSER: What about the metabolic effects of β blockers? Do they influence your choice? It is well known that the use of a β blocker will reduce HDL [high‐density lipoprotein] levels, increase tri‐glyceride levels, and increase insulin resistance—at least the β blockers without α‐blocking properties—like labetalol or carvedilol. Since we all pretty much agree that hypertension May be a metabolic disease, is that another reason to stay away from these drugs? How do you reconcile all this in the post‐MI patient and the results in high‐risk heart‐failure patients?
DR. CUSHMAN: I think where there is a clear indication for a β blocker, the metabolic effects should not be much of a consideration. On the other hand, where we have some evidence, as we have discussed, of some inferiority or no superiority for the β blocker compared with the other classes that we are dealing with, that is where the metabolic effects should be a greater consideration. Where you have a drug that May be less effective at preventing a coronary event or less effective at preventing a stroke, you certainly should consider the metabolic effects. On the other hand, a medication should not be chosen solely because it has a positive effect on metabolic parameters. An example of that would be an α blocker. An α blocker has better metabolic effects on lipids and glucose than most of the other drugs we use, but it clearly is inferior in terms of preventing cardiovascular events, as was demonstrated in ALLHAT.
DR. MOSER: Tom, how do you reconcile what we have been discussing with data from the United Kingdom Prospective Diabetes Study (UKPDS), where a β blocker‐/diuretic‐treated group was compared with an ACE inhibitor‐treated group? The fact was that both groups achieved the same dramatic decrease in all cardiovascular events, diabetic as well as nondiabetic, if the blood pressure was lowered to a great degree. This combination of a β blocker/diuretic would not appear to be a combination of choice in diabetics, but the results of treatment were excellent. Are the metabolic effects overwhelmed by the decrease in blood pressure?
DR. PICKERING: I think with diabetes there is a lot of evidence that the really important thing is the reduction of blood pressure. There are several studies that show that the lower the blood pressure, the fewer events; that is somewhat different from nondiabetic patients. Again, if β blockers are used in combination with diuretics, the benefits seem to still be present. I know, for instance, that in the Staesson meta‐analysis, there was no great superiority for the newer drugs when compared with β blockers and diuretics. I think we are focusing more on not using β blockers on their own rather than in combination with diuretics.
DR. MOSER: So your conclusion might be that β blockers should remain as part of the treatment for post‐MI and heart failure patients, and in combination it might be used with a diuretic as often as an ARB, ACE, or as a backup if an ARB or an ACE can't be used.
DR. PICKERING: Well, I would tend to favor an ARB or an ACE over the β blocker. There are other studies that do show a greater benefit. Also, I think that the β blockers May cause more side effects and fatigue than some of the newer drugs that also block the renin‐angiotensin system.
DR. CUSHMAN: I believe that one of the advantages of considering a β blocker in the anti‐hypertensive regimen of a patient who doesn't have a compelling indication for a β blocker is cost. However, we're getting to a point where ACE inhibitors are less expensive, and increasingly over the next few years, other classes of drugs will be. I agree with Tom. I much less often would use a β blocker as a second drug with a diuretic, even though that's what was done in studies like SHEP [Systolic Hypertension in the Elderly Program] and ALLHAT [Antihypertensive and Lipid‐Lowering Treatment to Prevent Heart Attack Trial], where the benefit for cardiovascular outcomes was clear with this combination. However, it is difficult to estimate how much, if any, benefit the β blocker added to the thiazide‐type diuretic had in those trials.
DR. MOSER: So you might use a β blocker as a third or fourth step if you have a really resistant patient.
DR. CUSHMAN: Perhaps, although I don't believe that β blockers add a lot to an ACE inhibitor or an ARB in terms of antihypertensive effect. So if you already have the ACE inhibitor or an ARB on board, I actually would prefer to add a CCB or reserpine, if needed, for blood pressure control.
DR. MOSER: Rather than another drug that's a weak renin inhibitor.
DR. CUSHMAN: Right. So I primarily use a β blocker either for compelling indications or when somebody has another condition such as migraine or a tremor, where the β blocker May have a beneficial effect.
DR. MOSER: The last issue is tolerability. Tom, you mentioned that β blockers May produce more side effects than the other commonly used antihypertensive drugs. Do you think that's true? I know in the MRC elderly trial a β blocker was not well tolerated, but that was an unblinded study. It seemed to me that there were more Raynaud's cases in that study than any one of us will see in 5 years. Perhaps since it was unblinded, if a patient's hands or feet were cold, the doctor automatically assumed that it was the β blocker and stopped it. If it had been a blinded study, they might have just waited to see what happened. What do you think? Do β blockers cause trouble?
DR. PICKERING: Well, that's a possibility, but as you said, the facts were that patients in the β blocker (atenolol) group did have more withdrawals. I think that when you compare them with the ARBs, for example, for which there are no consistently reported side effects, there's really no competition. There is some evidence that the ARBs May even reduce headaches in some studies. So I think they win hands down in terms of producing fewer side effects when compared with β blockers.
DR. MOSER: So the MRC study May have overemphasized adverse reactions, but most of us who have used β blockers have noted increased dreams, cold fingers, cold hands, and perhaps depression in some people. But what about β blockers with different activities? Most of the studies have been with propanolol, atenolol, or metoprolol. What about the β blockers with α‐blocking properties like labetalol or carvedilol, which do not increase vascular resistance and do not have significant deleterious effects on HDLs, triglycerides, and insulin resistance? Bill, how about carvedilol? Is that any better than atenolol?
DR. CUSHMAN: Let me back up a little. The irony is that in blinded studies such as the Treatment of Mild Hypertension Study (TOHMS) and the VA Single Drug Therapy study, β blockers were about as well tolerated as any of the other antihypertensive medications. Despite that, we have all seen adverse effects with β blockers that seem to be more frequent, at least to me, than with diuretics, CCBs, ACE inhibitors, and certainly ARBs.
DR. MOSER: OK, now what about a vasodilator β blocker like carvedilol? Or labetalol?
DR. CUSHMAN: I've participated in studies with both carvedilol and labetalol. They seem to be reasonably well tolerated; they certainly have fewer metabolic effects and, theoretically, better hemodynamic effects than the other β blockers. A limitation is that, except for heart failure, we really don't have any outcome studies with the vasodilating β blockers. Even though α blockade together with β blockade is really very effective at lowering blood pressure and lowers systolic blood pressure more than a β blocker alone, α blockade May cause more side effects.
DR. MOSER: Dizziness, postural hypotension…
DR. CUSHMAN: Yes.
DR. MOSER: Even with a β blocker on board?
DR. CUSHMAN: Even with a β blocker on board. Hypotension and dizziness are less common with labetalol and carvedilol than with an a blocker, but they are more common than with a β blocker as monotherapy. The biggest issue is that we really don't have any hypertension outcome trials.
DR. MOSER: Dr. Pickering, a last comment.
DR. PICKERING: I had the impression that with long‐term use, labetalol loses a lot of its α‐blocking effect and is not very different from propranolol. I think carvedilol is a genuinely different drug with unique properties and we shouldn't lump it in with the other β blockers. But we don't know much about its long‐term effects because it hasn't been used much in hypertension. Is it approved for use in hypertension?
DR. MOSER: Carvedilol has been approved, but has not been actively marketed for use in hypertension.
DR. PICKERING: It hasn't been advocated for hypertension as much as perhaps it should be.
DR. MOSER: So we can conclude that β blockers still have a role in the treatment of hypertension, but perhaps not as preferred monotherapy. In patients post MI, with angina and with heart failure, they are clearly indicated as part of therapy. In combination with a diuretic, they are highly effective and reduce morbidity and mortality, but May not be preferred when compared with an ACE inhibitor/diuretic or ARB/diuretic combination. Finally, a β blocker with α‐blocker properties with a more physiologic action and fewer adverse metabolic effects May be an effective substitute for a β blocker, but it would be nice to have outcome data in hypertensive patients for the α‐/β‐blocker class.