Abstract
Following a symposium on hypertension sponsored by the National Heart, Lung, and Blood Institute in New York City on January 17, 2001, a panel of the speakers was convened to discuss the topic, “Is it blood pressure lowering alone that accounts for the reduction in morbidity and mortality, or do specific medications or functions make a difference?” Moderating the panel was Dr. Marvin Moser, Clinical Professor of Medicine at Yale University School of Medicine. Participants included Dr. Thomas G. Pickering, Professor of Medicine at the Mount Sinai School of Medicine in New York and Dr. Michael A. Weber, Professor of Medicine at the State University of New York Health Science Center in Brooklyn, New York. Subsequent to this discussion, several treatment trials were reported. The results of these have been added where appropriate.
DR. MOSER: In 1996/97, when the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC) issued its guidelines and provided an algorithm for the treatment of hypertension, diuretics or β blockers were considered preferred initial treatment if lifestyle interventions were not successful in achieving a goal blood pressure of <140/90 mm Hg or even lower in patients with diabetes or renal disease. This suggestion was based on the results of 17 randomized clinical trials with these agents that had shown a reduction in morbidity and mortality. These trials had not addressed specific determinants of outcome other than blood pressure lowering. Did β blockers or diuretics have other effects or did they just reduce blood pressure, and was that enough to provide the most beneficial outcome?
Some investigators argued that enough attention was not being paid to specific effects of other agents that might produce even better results. In 1999, the World Health Organization (WHO) concluded that it was the blood pressure alone that made the difference; it didn't matter which drugs were given. They suggested that any one of five classes of drugs could be used as initial therapy with the expectation that equivalent results would be obtained. These included diuretics, β blockers, angiotensin‐converting enzyme (ACE) inhibitors, calcium channel blockers (CCBs), and α blockers. In view of the recent treatment trial results, is that conclusion reasonable? Are the WHO recommendations more reasonable than the JNC recommendations of 1997? Should we consider changing the JNC guidelines or, on the other hand, should the WHO recommendations be modified?
DR. PICKERING: My own view is that we should probably modify the JNC VI recommendations a little, to add ACE inhibitors as a third category for first‐line treatment. I would not favor adding α blockers or CCBs as initial monotherapy, except for CCBs in elderly hypertensives; the JNC VI already suggested this, as an alternative to a diuretic in subjects with isolated systolic hypertension.
DR. MOSER: Dr. Weber, what about that? Do you agree, and what are the newer data to justify this change?
DR. WEBER: I have always liked the guidelines that were put out by the WHO International Society of Hypertension (ISH) group, which basically took the view that any of the available drug classes was acceptable for initial therapy—this despite some pros and cons. Any medication would be useful, provided that it worked to reduce blood pressure and was well tolerated by patients. I agree with what Dr. Pickering said: certainly, the evidence with ACE inhibitors has now accumulated to a point where it's hard to ignore them as attractive first‐line treatments. I also believe that it's important to say early in this discussion that a majority of patients are going to be on more than one antihypertensive drug; therefore, what we have to consider is that we would like to see an ACE inhibitor or a diuretic or β blocker included in a person's regimen, but it does not necessarily have to be the first drug prescribed.
DR. MOSER: So the whole debate about which drug to use as initial therapy may be moot.
DR. WEBER: Well, it's moot to a large extent, though this issue is not yet fully resolved. There is now growing acceptance that lowering blood pressure per se is a very important objective, and may be the most important aspect of cardiovascular protection. If patients reach acceptable levels of blood pressure, however this is achieved, most clinicians are probably going to leave them alone unless they have other compelling problems, such as diabetes, to justify a re‐evaluation of their therapy.
DR. MOSER: Now, the WHO‐ISH guidelines were issued in 1999, so they had 3 more years of trial data to consider than the JNC VI group. But even they speculated about some of their conclusions; some of the comparative trials weren't finished yet. I agree with you that at the present time, ACE inhibitors should be added to diuretics and β blockers as possible initial therapy. What data do we have to substantiate this recommendation? Also, do we have any data that agents that block the renin‐angiotensin system should be used preferentially in hypertensive subjects with comorbid conditions?
DR. PICKERING: Yes, there are several studies that have been published since the WHO‐ISH guidelines, some of which have compared ACE inhibitor‐based treatment with β blocker‐ and diuretic‐based treatment. These indicated that the two treatment approaches are equivalent in terms of their reductions of cardiovascular morbidity. Specifically, the Swedish Trial in Old Persons (STOP 2) indicated that an ACE inhibitor‐based treatment program reduced cardiovascular events and blood pressure to as great a degree as a diuretic‐ or β blocker‐based regimen. In addition, it appeared that the incidence of myocardial infarction and heart failure was reduced to a greater degree with ACE inhibition than with CCB treatment, despite a similar decrease in blood pressure. So it appears that, at least in the CCB/ACE inhibitor comparison, there may have been a difference in outcome with different drugs.
DR. MOSER: The other comparative studies of interest include the United Kingdom Prospective Diabetes Society (UKPDS) trial. In this trial of hypertensive type 2 diabetics, equivalent results were obtained with ACE inhibitor‐ and β blocker‐based treatment. Fewer events occurred if tight blood pressure control was maintained, compared to a group of patients with less effective blood pressure control (a difference of −10/−5 mm Hg between groups). The blood pressure differences, not the type of therapy, appeared to account for the better outcome. In another trial, however, the Captopril Prevention Project (CAPPP), overall cardiovascular events were reduced equally when an ACE inhibitor group was compared to a diuretic/β blocker group, but there were fewer events in diabetics in the ACE inhibitor group. This conclusion is somewhat different from the UKPDS results.
Nevertheless, all of these trials have indicated that ACE inhibitor‐based therapy is equal to, or perhaps in some subgroups more effective than, diuretic/β blocker‐based therapy in reducing cardiovascular events.
It should be remembered, as Dr. Weber pointed out, that not all of the studies were trials of monotherapy. More than 50% of subjects were on two or more drugs.
DR. PICKERING: And then there was the Heart Outcomes Prevention Evaluation (HOPE) study, where an ACE inhibitor was added to conventional therapy in patients who were at particularly high risk for cardiovascular disease. In that study, there was a substantial reduction of risk, which was much greater than would have been anticipated by the rather modest reduction in blood pressure.
DR. MOSER: Of course, the HOPE study was not a hypertension study. Only about 46% of subjects had hypertension; they were very high‐risk patients—more than 80% had cardiovascular disease. These are not exactly the kinds of patients you're going to see every day.
DR. WEBER: I agree with that statement, but in many ways, HOPE was a hypertension study. When you look carefully at the data presented in the different blood pressure subgroups, there was no question that the high‐risk patients who came into the HOPE trial with hypertensive blood pressure levels had by far the greatest benefit, even when the blood pressure was not reduced to normal levels. Hypertensive patients appear to be at greater risk than can be explained by high blood pressure alone. We've got a lot more to learn about hypertension; I suspect there is something beyond the blood pressure that makes hypertensive patients so vulnerable to vascular disease and that, at least in some instances, specific therapy does make a difference.
DR. MOSER: But you did say that as far as you are concerned, the major benefit is achieved by reducing the blood pressure.
DR. WEBER: Suppose the hypertensive patients in the HOPE trial had been treated with a non‐ACE inhibitor drug and had become normotensive. Would they have gotten the same benefit that they appeared to derive from ACE inhibitor therapy?
DR. MOSER: Okay, we had β blocker/diuretic data. Then some good data from the UKPDS and STOP 2 studies were published, showing that ACE inhibitors with or without diuretics were as effective as β blockers and diuretics. Additional data from several other trials, such as CAPPP, also indicated that ACE inhibitor‐based treatment lowered blood pressure, and that morbidity and mortality were reduced. So I agree; we can certainly say that the benefit of using an ACE inhibitor appears to be equivalent to that achieved with diuretics and β blockers.
What about CCBs? These are still the second most widely used antihypertensive medications in the United States. What data do we have on these agents to suggest that they are as effective as the other agents? Should they also be considered as first‐step therapy? The JNC VI recommendations suggest that a long‐acting dihydropyridine CCB is an appropriate alternative to a diuretic in patients with isolated systolic hypertension. This was based on results of the Systolic Hypertension in the Elderly trial in Europe (Syst‐Eur), which demonstrated a reduction in strokes and overall cardiovascular events in the CCB group compared to the placebo group. Dr. Pickering, you said that at the moment, on the basis of available data, you would not recommend CCBs as initial monotherapy, except in some situations, despite their effectiveness in lowering blood pressure.
DR. PICKERING: Well, the CCB story has been a controversy for several years. I think what is becoming clear is not that they are dangerous, except for possibly the very short‐acting ones that we don't use any longer, but that they appear to be less effective than, for example, ACE inhibitors in reducing coronary heart disease events and congestive heart failure. In most patients, coronary events are the most important risk. That would be one reason why I would not routinely use them as first‐line treatment.
But having said that, there are many patients who require multiple drugs for adequate blood pressure control, and adding a CCB may be a very effective way of achieving that. Also, there are good data indicating that they are as effective as other agents, or more so, in reducing strokes.
DR. MOSER: So CCBs lower blood pressure, and they're reasonably well tolerated, but the data appear to indicate that while they may be as effective in reducing strokes, they may not reduce coronary heart disease events and heart failure to as great a degree as other agents. Dr. Weber, is that a fair statement?
DR. WEBER: Yes, it is. I feel reassured that these agents are safe and effective—not only by the Syst‐Eur trial for elderly people with predominantly systolic hypertension, but also by the results of the Intervention As a Goal in Hypertensive Treatment (INSIGHT) trial with another dihydropyridine. The overall results in this trial showed similar benefits with the CCB therapy and conventional treatment. The Nordic Diltiazem (NORDIL) trial, which evaluated a non‐dihydropyridine, reported a slightly better outcome for stroke, and perhaps slightly less benefit for myocardial infarction, with the CCB.
I'm left with the feeling that if we are going to recommend drugs as first‐line therapy, there probably isn't a great deal of difference among them except to go back to what JNC VI did in its last report: utilize the concept of “compelling indications.” For instance, the presence of diabetes would incline us toward an ACE inhibitor or an angiotensin receptor blocker (ARB). People at risk of heart failure or already in heart failure would also require an ACE inhibitor—with a diuretic, of course.
DR. MOSER: Getting back to what Dr. Pickering said, the greatest risk in terms of numbers, not in terms of disability or tragedy, is in coronary artery disease events. In the INSIGHT study, for example, although cardiovascular events were equally reduced overall with a dihydropyridine and a diuretic, there was an increased incidence of myocardial infarction and heart failure in the CCB group. This does not suggest that the CCBs increase coronary heart disease events, but merely that compared to diuretics, CCBs may decrease these events to a lesser extent. It appears that when you compare ACE inhibitors to CCBs or CCBs to diuretic‐based treatment programs, there may be a greater benefit in outcome in terms of myocardial infarctions and congestive heart failure but not in strokes with agents other than CCBs. That may be why Dr. Pickering suggested that at present they're not to be placed in the category of first‐line agents. I agree. CCBs clearly have a place in treatment; they are effective antihypertensive drugs but they may not be quite as beneficial as the other drugs in terms of outcome. So we are saying that while lowering blood pressure may be the most important aspect of reducing morbidity and mortality, there may be different results with different agents.
DR. PICKERING: I think one additional thing that keeps emerging is that diuretics still look very good in terms of both treating hypertension and their effects on morbid events. It is also important to note that that they can add to the blood pressure‐lowering effect of just about any of the other antihypertensive drugs.
DR. MOSER: Well, let's get back to what Dr. Weber said about compelling reasons, or specific reasons, to use a specific medication. Probably an ACE inhibitor or an ARB should be part of the regimen in diabetics, especially those with nephropathy. But in all of the trials in diabetics, more than half of the patients were on at least one other drug. So should we be suggesting that first‐step therapy in a diabetic should comprise, for example, an ACE inhibitor or an ARB and a diuretic, or a β blocker plus a diuretic (based on the UKPDS study)? Or are we not yet ready to recommend that?
DR. WEBER: If you're thinking about the importance of achieving aggressive blood pressure goals, I think that makes a great deal of sense because the goals of therapy are lower in diabetics than in other hypertensives.
DR. MOSER: What are they now, Dr. Weber? Haven't they been changed recently?
DR. WEBER: The most recent recommendation of the American Diabetes Association is to get the blood pressure down to 130/80 mm Hg in diabetics, even in those who do not have evidence of overt kidney disease. These are aggressive new guidelines. Very rarely are we going to reach that target with the one drug; whatever the reason, the blood pressure is often more difficult to control in diabetics than in nondiabetics.
DR. MOSER: Especially if they have proteinuria or renal disease of any significance.
DR. WEBER: Exactly. You're going to end up with multidrug therapy. What seems to make things a little easier is that there is now growing acceptance that when using a diuretic in diabetic patients, it's reasonable to stay with traditional thiazides. We had previously believed that it was necessary to start using loop diuretics as soon as the creatinine reached somewhere above 1.8 mg/dL, but this now appears not to be strictly necessary. You may have to use a higher than usual dose of a thiazide to get good results, but since the thiazides are better tolerated and so much easier to use than the loop diuretics, it's probably worthwhile to persist with them. Bear in mind, however, that doses of 25 or 50 mg, or an equivalent dose of hydrochlorothiazide, may have to be used in these patients.
DR. MOSER: Do you believe that any medication that reduces the diabetic's blood pressure below 130–135/80–85 mm Hg is going to be just as effective as any other? In other words, can we differentiate among the drugs?
DR. WEBER: Recent preliminary presentations at the American Society of Hypertension indicated that ARBs may produce benefits beyond their blood pressure effects in diabetics with early or established nephropathy. There was also evidence—for example, in the African American Study of Kidney Disease (AASK) trial—that in patients who already have nephropathy, a dihydropyridine may not be as effective as a β blocker or an ACE inhibitor in providing renal protection.
DR. MOSER: What was that study? Describe it for us.
DR. WEBER: That was the AASK trial. A β blocker (Toprol‐XL®) or an ACE inhibitor (ramipril) were similarly effective in preventing further deterioration of renal function, but the dihydropyridine CCB amlodipine was not as effective in preserving renal function. These medications had been added to other therapies in a randomized fashion in more than 1000 African American patients with hypertension and renal disease.
DR. PICKERING: But that was only in people who had definite proteinuria to begin with.
DR. MOSER: We have these preliminary data from the AASK trial, and we have data that are somewhat controversial from the Fosinopril‐Amlodipine Comparative Study (FACET), and from the Appropriate Blood Pressure Control in Diabetes (ABCD) trial. The last two studies compared the effects of an ACE inhibitor and a dihydropyridine CCB. These trials reported that ACE inhibitors are superior to CCBs not only in decreasing the progression of renal disease but also in reducing myocardial infarctions and heart failure. These are relatively small studies, and have been criticized, but they seem to present a pattern which, if we were to back up and look at the data objectively, would lead to a conclusion that ACE inhibitors are probably a better first‐choice drug in diabetics, especially those with proteinuria, than CCBs. This probably also applies to ARBs, in light of new data. Is that fair?
DR. WEBER: I think that's a fair statement, particularly in that population of diabetics, but I'd emphasize that even though the authors of FACET have not highlighted it, the third group of patients in that trial, who received both an ACE inhibitor and a CCB, appeared to have the best event reduction in of all. This emphasizes your earlier point, Dr. Moser, that first of all we must get the blood pressure down as much as possible and then worry about what drugs are being used.
DR. MOSER: So if we had to summarize the relative importance of blood pressure lowering and different therapies in reducing morbidity and mortality, we might conclude that most of the benefit is the result of blood pressure lowering, but perhaps in certain circumstances, the benefit may be greater with specific medications. For example, a regimen based on an agent that blocks the renin‐angiotensin system—an ACE inhibitor or an ARB—may be more effective in certain situations than a CCB‐based regimen, or a diuretic may be more effective than a dihydropyridine in reducing coronary heart disease events. Of course, there is no way of quantifying what percentage of benefits accrue from blood pressure lowering alone; probably, most of the benefit is from better blood pressure control. The Table summarizes some of the similarities and differences in outcomes among the various drugs.
Table.
Some Medications May Have Specific Advantages Over Others in the Management of Hypertension: Comparative Evidence
| STOP‐2 in Elderly: Fewer MIs and less CHF with ACE inhibitors than with a dihydropyridine CCB |
| CAPPP: Fewer CV events in diabetics with ACE inhibitors than with a β blocker/diuretic regimen |
| ALLHAT: Fewer CV events with a diuretic than with an α blocker |
| INSIGHT: Fewer MIs and less CHF with a diuretic than with a dihydropyridine CCB |
| NORDIL: Fewer strokes, but a tendency toward more CHD events with a nondihydropyridine CCB than with a β blocker/diuretic regimen |
| FACET, ABCD, and AASK: Fewer CV events and less progression of renal disease with an ACE inhibitor‐based than with a CCB‐based treatment program in hypertensive diabetics with renal disease or hypertensive African Americans with renal disease |
| RENAAL, IRMA II, and IDNT: Slowing of renal disease progression in type 2 diabetic hypertensives with an ARB (usually with other medication), compared to patients not receiving an ARB, ACE inhibitor, or CCB |
| MI=myocardial infarction; CHF=congestive heart failure; ACE=angiotensin‐converting enzyme; CCB=calcium channel blocker; CV=cardiovascular; CHD=coronary heart disease; ARB=angiotensin II receptor blocker ALLHAT=Antihypertensive and Lipid‐Lowering Treatment to Prevent Heart Attack Trial; RENAAL=Reduction of Endpoints in NIDDM with the Angiotensin II Antagonist Losartan; IRMA=Irbesartan Microalbuminuria Type 2; IDNT=Irbesartan Type 2 Diabetic Nephropathy Trial (other acronyms are spelled out in the text) |
DR. PICKERING: I would agree with that, overall. I think one of the things we have learned is that there appear to be specific subsets of patients, such as patients with renal disease and proteinuria, in whom one type of agent is definitely preferable to another. I think the bottom line is still to get the blood pressure down by whatever means and with whatever combination that works.
DR. MOSER: Right. We all agree that about 40%–50% (we don't know the exact number, but that is a good guess) of people are going to respond to monotherapy regardless of the medication used. It is probably less than this in diabetics, patients with renal disease, and perhaps in the elderly. So how do we convince physicians to move away from the concept of always starting with monotherapy and up‐titrating? It would be useful if we had a method to determine, beforehand, who will or will not respond to a specific agent, but, despite some claims to the contrary, we do not. So we are faced with using at least two different medications from different classes, either separately or as a combination. The latter approach has some advantages. There are now several studies demonstrating that a combination that contains a diuretic not only will lower blood pressure more effectively but probably will result in a better outcome than a regimen that doesn't include a diuretic. Results of the Swedish Study in Old Persons indicate that in the ACE inhibitor group, in which there were fewer myocardial infarctions, more of the patients received a diuretic as a second drug than in the CCB group, where a β blocker was often the second drug.
How do we convince physicians that multiple drug therapy is an option? Would you start a patient without renal disease or without diabetes on combination therapy, or two different medications (not necessarily in combination)? Is it reasonable to go that far now? The JNC said it was appropriate, so should we be even more emphatic?
DR. PICKERING: I treat quite a lot with low‐dose combinations of β blockers and diuretics. I think that this is a very tried and effective combination that's usually quite well tolerated.
DR. MOSER: I agree; a combination of an ACE inhibitor or an ARB and a diuretic are also effective and well tolerated. Dr. Weber?
DR. WEBER: I think fixed combinations are going to be a big part of the future of treating hypertension, and they certainly allow us to get more rapid and more effective control of blood pressure. This may be important in giving patients confidence that we are taking appropriate steps to bring the blood pressure under control.
DR. MOSER: And they think you're smarter if you get it right the first time.
DR. WEBER: That's right. There are definitely rewards for quick success. But I suspect that combinations with or without a diuretic will be effective and I believe that even some of the ACE inhibitor/CCB fixed combinations have a lot to offer. The point is that logical combinations work pretty well, and are also well tolerated.
DR. MOSER: What about the ARBs? Where do they fit, and do they have any advantages over other therapy? Are the recent studies definitive? Should we use ARBs only if someone develops a cough from an ACE inhibitor? Are there any reasons to use them interchangeably with an ACE inhibitor as initial therapy?
DR. PICKERING: Or do we use them with an ACE inhibitor? There's a recent, intriguing study from Denmark about combining ARBs with ACE inhibitors in diabetics, in which the combination not only improved blood pressure control, it also reduced proteinuria more than either agent alone. Many of us have traditionally thought that since the mechanism of action with these two agents is somewhat similar, this might not be a good combination. It could be that we were wrong.
DR. WEBER: And of course the combination is helpful not only in diabetic nephropathy, but their use may also be beneficial in heart failure. In the Valsartan Heart Failure Trial (VAL‐HEFT), patients on an ACE inhibitor plus other therapy for heart failure were randomized to receive either placebo or a high dose of an ARB. There was a strong suggestion of benefit, particularly a reduction in hospitalizations.
DR. MOSER: In that study, people on an ACE inhibitor, as well as a β blocker, and in most cases, digitalis and a diuretic actually appeared to do worse, as far as mortality was concerned. So perhaps an ARB shouldn't be given to patients with heart failure who are already on an ACE inhibitor or β blocker.
DR. WEBER: But what is of interest is that in those patients who were not already on an ACE inhibitor, and there were a number who, for whatever reason, did not stay with the ACE inhibitor, the ARB, either alone or in combination with the β blocker, was very effective. So I think we can be pretty confident that the ARBs are at least going to rival the ACE inhibitors in terms of preventing major clinical outcomes in heart failure patients.
DR. MOSER: We do have evidence that these agents decrease blood pressure, decrease proteinuria, and reverse left ventricular hypertrophy to about the same degree as the ACE inhibitors.
If we were to write JNC VII today, do we all agree that we would add ACE inhibitors to diuretics and β blockers as possible initial therapy and emphasize combination therapy with an ACE inhibitor/diuretic or a β blocker/diuretic as initial therapy?
DR. PICKERING: Yes, I think so.
DR. MOSER: If we were rewriting JNC VI, and if we stick to evidence‐based medicine, we would not be able, at this time, to make a strong statement about the ARBs, except in the management of hypertensives with type 2 diabetes. But as you know, the WHO‐ISH recommendations in 1999 suggested that any of the classes of drugs that lower blood pressure are acceptable as initial treatment. Is it fair to conclude today that an ARB plus a diuretic is probably interchangeable with an ACE inhibitor plus diuretic combination?
DR. PICKERING: I would qualify that statement slightly. There are ongoing trials, which, if the results are positive, would give us the confidence to recommend that ARBs could be seen as interchangeable with ACE inhibitors. It would certainly be reasonable at present to use ARBs in diabetic patients or those in whom, for whatever reason, one doesn't want to use an ACE inhibitor.
DR. MOSER: To repeat, this discussion about initial therapeutic preferences may be moot because we may be moving toward a conclusion that the use of two drugs initially may be the way to begin therapy, especially in higher‐risk subjects. A (3 blocker, ACE inhibitor, or ARB with a diuretic, or in some cases, a CCB with a diuretic, are highly effective combinations. An ACE inhibitor/CCB is also an alternative. Despite the fact that some investigators still believe that a CCB plus a diuretic is not more effective than monotherapy with either agent alone, there are data to the contrary.
DR. PICKERING: That's a good thought, Dr. Moser. That would be a combination well worth exploring in a number of ways.
DR. MOSER: We agree that most of the benefit in reduction of mortality and morbidity in hypertensive patients results from lowering blood pressure, but that there may be some differences among medications.
-
•
It appears that the older medications, i.e., diuretics and β blockers, are as effective as the newer agents in both the elderly and diabetics. (There is some evidence that an ACE inhibitor‐based program—and very likely an ARB regimen as well—is more effective in diabetics.)
-
•
There is evidence that an ACE inhibitor‐based treatment program lowers the incidence of myocardial infarction and heart failure to a greater degree than a dihydropyridine CCB in diabetics.
-
•
Overall, cardiovascular events appear to be reduced equally by a diuretic and either a dihydropyridine or nondihydropyridine CCB, but there appear to be fewer myocardial infarctions and episodes of congestive heart failure when a diuretic is given. One trial suggests that there are fewer strokes with a nondihydropyridine CCB than with a diuretic, and a recent meta‐analysis seems to support the idea that CCBs may be as effective as drugs of other classes in preventing strokes. Some data indicate fewer strokes when a diuretic regimen is compared to studies where a diuretic is not used.
-
•
A regimen that includes an ARB is more effective in slowing renal disease progression in type 2 diabetics than a treatment program that does not include an ACE inhibitor, ARB, or CCB.
-
•
Finally, recent data reaffirm that the use of small doses of two different medications may be an appropriate first step in treatment.