Author + information
- Received November 7, 1996
- Revision received February 13, 1997
- Accepted February 26, 1997
- Published online June 1, 1997.
A review of recent studies suggests that the use of angiotensin-converting enzyme (ACE) inhibitors may be preferred (usually along with a diuretic drug) as initial therapy in several subsets of hypertensive patients (i.e., those with diabetes and nephropathy or with diminished left ventricular function with or without symptoms of heart failure). Limited long-term data are available for the angiotensin II receptor antagonists. The use of nondihydropyridine calcium channel blocking agents (CCBs) appears to reduce reinfarction in patients with ischemic heart disease (however, mortality is not reduced). Long-acting formulas of CCBs appear to decrease congestive heart failure in patients with dilated, but not ischemic, cardiomyopathy and to decrease strokes and arrhythmias in hypertensive subjects. Short-acting agents (primarily those that increase heart rate) may increase coronary heart disease events in hypertensive patients. There is little evidence at present that CCBs offer a major advantage over other antihypertensive agents or that they should be recommended as initial therapy, except in special situations.
(J Am Coll Cardiol 1997;29:1414–21)
In 1993 the Fifth Report of the Joint National Committee on Detection, Evaluation and Treatment of High Blood Pressure (JNC V) () in the United States changed recommendations from their previous report () for initial therapy of hypertension. In 1988 (), this committee had suggested a choice of diuretic drugs, beta-adrenergic blocking agents, angiotensin-converting enzyme (ACE) inhibitors or calcium channel blocking agents (CCBs) as initial treatment. In contrast, in 1993 (), the use of diuretic drugs or beta-blockers was recommended as preferred initial therapy and the use of ACE inhibitors, CCBs, alpha-blockers and alpha/beta-blockers as possible alternative therapy (Table 1). The committee was criticized for this recommendation (); however, their recommendation reflected the evidence from long-term clinical trials. At that time, diuretic drugs and beta-blockers were the only two classes of drugs that had been tested and shown to reduce morbidity and mortality in hypertensive subjects (). If the data are examined carefully, the case for diuretic drugs as initial therapy is stronger than that for beta-blockers, especially in the elderly. The Medical Research Council (MRC) trial (), for example, had failed to show a benefit with regard to stroke reduction with beta-blockers, but several of the trials had demonstrated a reduction in myocardial infarction (MI) when these agents were used, and the long-term use of beta-blockers in patients with ischemic heart disease has been shown () to reduce the incidence of reinfarction. In studies such as the Systolic Hypertension in the Elderly Program (SHEP) (), a beta-blocker was added to a diuretic drug if goal blood pressure was not achieved with monotherapy. However, most of the achieved reduction in cardiovascular disease events were attributable to the use of diuretic drugs. The JNC V suggested the use of ACE inhibitors or CCBs in special situations or where patients were unable to tolerate the “preferred initial treatment.” Since the JNC V report, a new class of medication, the angiotensin II (A-II) receptor antagonists, which block the effects of the renin–angiotensin system at a peripheral site, has been approved for treatment of hypertension (Fig. 1). It is of interest to examine recent reports regarding the ACE inhibitors, A-II receptor antagonists and CCBs to determine whether the data are sufficient for these drugs to be considered as preferred therapy in the management of hypertension.
1 The Case for ACE Inhibitors
ACE inhibitors have been extensively studied and have been shown in double-blind, placebo-controlled studies ([8, 9]) to reduce blood pressure to the same extent as beta-blockers and to nearly the same degree as diuretic drugs in whites but not in blacks. ACE inhibitors are well tolerated; side effect studies and quality of life measures have demonstrated that they are tolerated as well or better than other medications. Recent double-blind, quality of life measures () have not demonstrated that ACE inhibitors are superior to diuretic drugs or at least one type of beta-blocker.
2 ACE Inhibitors in Diabetes and Renal Disease
There are subsets of patients in whom the ACE inhibitors appear to offer significant advantages. Because of the selective effect of ACE inhibition on the afferent arteriole of the glomerulus and a reduction in arteriolar resistance, the use of these agents has proved beneficial in the treatment of patients with proteinuria and diabetic nephropathy (). Some questions must be asked: 1) Are the ACE inhibitors unique? 2) Do they have a specific effect in protecting the kidney in patients with diabetes or other forms of nephropathy? 3) Is the benefit achieved solely as a result of blood pressure lowering? Studies by Parving et al. in 1983 () and 1987 () reported that in patients with insulin-dependent type I diabetes, with evidence of nephropathy, treatment with a beta-blocker (metoprolol), a vasodilator (hydralazine), a centrally acting agent (methyldopa), a diuretic drug and a peripheral sympathetic blocking agent (quanethidine) reduced the slope of decline in glomerular filtration rate as blood pressure was lowered. Other studies () have shown that the use of any antihypertensive agent will decrease proteinuria if pressure is lowered, but agents that block the renin–angiotensin system may be especially effective.
The ACE inhibitors and more recently an angiotensin receptor antagonist, losartan, have been extensively studied in patients with type I diabetes and nephropathy ([10, 14, 15]). In a multicenter, randomized trial of type I diabetes (), an ACE inhibitor was compared with placebo in a double-blind study. However, in the ACE inhibitor group, ≥75% of patients also required diuretic drugs and beta-blockers to achieve normotensive blood pressure levels in the hypertensive subjects. The placebo-group patients also received diuretic drugs and beta-blockers but did not receive an ACE inhibitor. In the ACE inhibitor group there was a significant reduction in the number of patients who progressed to severe or end-stage renal disease or death and a 50% increase in the time it took for the creatinine level to double. Other data suggest that blockade of the renin–angiotensin system will decrease proteinuria. It has been suggested () that these drugs be used preferentially (usually with a small dose of a diuretic drug) not only in type I diabetic hypertensive patients with nephropathy but also in patients with type II diabetes in an effort to slow down the progression of renal disease. This recommendation appears to be a reasonable one.
Studies with the A-II receptor antagonists are limited but also suggest a benefit of treatment in patients with proteinuria and nephropathy ().
In suggesting that ACE inhibitors or A-II receptor antagonists be used preferentially in patients with renal disease, it is important to remember that renal function may be worsened and acute renal failure precipitated in patients with creatinine levels >2.5 to 3.0 mg/dl and that hyperkalemia may occur in these patients when these drugs are used. In addition, ACE or A-II receptor inhibition reduces the elaboration of aldosterone, which in time may lead to potassium retention. Most investigators advise caution in using these medications in patients with creatinine levels >3 mg/dl.
3 A-II and Aldosterone Inhibition in Heart Failure
Another area where abundant data have been collected with regard to the effects of inhibition of the renin–angiotensin system is in patients with congestive heart failure (CHF), especially secondary to systolic dysfunction ([16–19]). There is good evidence of a hormonal effect in CHF, with a marked increase in A-II and aldosterone levels with vasoconstriction and sodium retention. It appears logical, therefore, that blockade of the renin–angiotensin–aldosterone system, whether at the level of ACE inhibition or at the A-II receptor site, would be of benefit.
Several studies ([16–18]) have shown that patients with CHF, either symptomatic or asymptomatic, and an ejection fraction <35% who received an ACE inhibitor showed a statistically significant reduction in mortality compared with patients who did not receive an ACE inhibitor. In these studies digitalis, diuretic drugs and a low sodium diet were also given in most cases. In patients treated with an ACE inhibitor plus “usual” management, a reduction in hospital admissions for CHF as well as MI was also noted. In a subset of hypertensive patients in one of these studies (Studies of Left Ventricular Dysfunction [SOLVD] []), there was a significant reduction in CHF but not in the occurrence of angina, strokes or MI (). However, in normotensive patients, a reduction in angina, MI and total mortality was noted (Table 2).
In another study (SAVE) (), patients with an MI and an ejection fraction <40% in the postinfarction period also benefited significantly when an ACE inhibitor was given in addition to routine therapy. Deaths from cardiovascular causes, recurrent MI and CHF were reduced. Some of these patients had definite evidence of CHF, whereas some did not.
A meta-analysis () of 16 randomized trials reported that reduction in total mortality was dependent on pretreatment ejection fraction. In patients with an ejection fraction <25%, there was a significant reduction in total mortality rate of ∼31%. However, this trend was not seen in patients with an ejection fraction >25%. From these data, it appears that the subset of patients with a low ejection fraction and systolic dysfunction, with or without a previous MI, will benefit from inhibition of the renin–angiotensin–aldosterone system. Life is prolonged, recurrent CHF is reduced, and, in some cases, MI can be prevented.
It is well established that the use of ACE inhibitors, as well as diuretic drugs, beta-blockers and CCBs, will result in a reduction in left ventricular (LV) mass and wall thickness in hypertensive patients.
Studies with A-II receptor antagonists are somewhat limited, with no data on long-term reduction of mortality and morbidity in hypertension. The use of losartan, the A-II receptor that has been most extensively studied, has been shown () to produce beneficial vasodilator, hemodynamic and neurohormonal effects in heart failure over a 3-month study period. In addition, a recent 48-week study () in elderly patients with CHF reported a significant reduction in overall mortality, primarily due to a decrease in sudden deaths, with losartan versus an ACE inhibitor captopril. More than 75% of patients in both treatment groups were receiving a diuretic drug or digitalis, or both.
The case for advising the preferential use of ACE inhibitors in subsets of hypertensive patients with proteinuria and diabetes, with or without evidence of nephropathy, and in subjects with reduced LV function, with or without classic signs of CHF, appears to be a good one. As noted later, a strong case can be made for using these agents in combination with small doses of a diuretic drug. Definitive outcome data are not yet available for the A-II receptor antagonists. We must await long-term trial results before a specific recommendation for the use of these medications as preferred initial therapy in other hypertensive subjects can be made. On the basis of available data on effectiveness and tolerability, it is probable that ACE inhibitors will be shown to reduce morbidity and mortality.
4 The Case for CCBs
Is there evidence that the use of CCBs reduces morbidity or mortality in hypertensive subjects or in subsets of patients described as benefiting from ACE inhibitors?
To date there are no well controlled, randomized, long-term studies with these agents in hypertensive patients. CCBs lower blood pressure to a degree equivalent to other medications, and the long-acting formulations are generally well tolerated by most patients (). But do they have unique properties to warrant a recommendation for widespread use while awaiting the results of ongoing trials? Animal studies had demonstrated a slowing of the progression of atherogenesis and a reduction of post-MI muscle injury with CCBs. It was theorized that the reduction of calcium ion influx into smooth muscle cells with these agents would reduce smooth muscle growth and vasoconstriction and reduce plaque formation. As a result of heavy promotional efforts, the CCBs were accepted as initial therapy throughout the United States and the world for the treatment of hypertension and angina. However, these animal studies had been performed, in most cases, with dosages from 20 to >40 times the dosages used in humans.
Although CCBs are effective antihypertensive drugs and, as noted, have been shown in several double-blind, controlled studies to reduce blood pressure to a degree equal to that of diuretic drugs, beta-blockers and ACE inhibitors, as well as A-II receptor antagonists, there are some disturbing data that suggest that the widespread use of these agents should be reexamined ([22–26]). Although the use of ACE inhibitors has been demonstrated to benefit patients with CHF or a previous MI, or both ([16, 17]), the evidence of benefit with CCBs in subjects with ischemic heart disease is equivocal or negative.
It should be emphasized that CCBs comprise a heterogenous group of drugs and are available in formulations (rapid or slow release) that may result in different physiologic and clinical outcomes.
5 Effects of CCBs in Patients With Ischemic Heart Disease
Studies in patients with angina have reported that the use of the short-acting, or rapid-release, dihydropyridine CCBs (mainly nifedipine) does not reduce the incidence of recurrent MI, CHF or mortality. Several studies ([27, 28]) of the nondihydropyridine CCBs (diltiazem and verapamil), whose use does not result in an increased heart rate, have shown some reduction in recurrent MI but not in patients with evidence of pulmonary congestion. When results in patients with and without CHF are pooled, the combined results show no statistically significant reduction in overall post-MI events.
A meta-analysis of the effects of verapamil and diltiazem after MI showed that reinfarction rates were reduced by ∼20 to 22% (). A meta-analysis () of studies with nifedipine suggests that this short-acting dihydropyridine actually increased mortality in patients with ischemic heart disease. Table 3summarizes the results of randomized trials of various CCBs in patients with an MI or unstable angina.
A recent study () of the long-acting dihydropyridine CCB amlodipine in patients with diminished LV function demonstrated a benefit in subjects with dilated cardiomyopathy but no benefit in those with ischemic cardiomyopathy. There was no reduction in mortality in the CCB-treated group. Other studies with nondihydropyridine CCBs had reported similar findings.
6 Effects of CCBs on Progression of Atherosclerosis
Several placebo-controlled studies have examined angiographic changes in patients receiving the short-acting dihydropyridine nifedipine (INTACT study) () or nicardipine (Montreal Heart Study) (). Neither of these studies showed angiographic evidence of a difference in progression or regression of lesions compared with use of placebo. New lesions were reduced somewhat, but overall the studies did not confirm animal data suggesting that atherogenesis could be slowed by CCB therapy.
A recent prospective, 3-year comparative study () of another short-acting dihydropyridine (isradipine) in hypertensive subjects was designed to show whether this agent reduced the progression of intimal–medial thickness in the carotid artery to a greater degree than hydrochlorothiazide. This method of evaluation of artery lesions closely parallels the changes in coronary arteries on angiography. At the end of 3 years, a decrease in blood pressure was noted with both agents; the decrease in systolic pressure was greater with hydrochlorothiazide (−19.5/−13 vs. −16/−13 mm Hg). The investigators had expected that the use of hydrochlorothiazide, because of its reported adverse effects on cholesterol levels and insulin resistance, might result in a greater degree of progression of atherosclerosis than with isradipine. However, at the end of 3 years, there was no difference in the rate of progression between the two drugs. A disturbing finding, which was also noted in the INTACT () and Montreal Heart Study () trials, was that the number of cardiovascular events was actually increased in the isradipine group (25 vs. 14 in the hydrochlorothiazide group) and that the incidence of new-onset angina was increased in the dihydropyridine group (Table 4).
Thus, data on the short-acting dihydropyridine CCBs with regard to progression or regression of atherosclerosis appear to be negative. Of importance to clinicians is that in each of the studies thus far reported, a troublesome increase in cardiovascular events has been noted. There is also little evidence that the use of CCBs reduces the incidence of MI or the recurrence of events in patients with previously diagnosed ischemicheart disease, except in subsets of patients without evidence of CHF who receive nondihydropyridine formulations.
In one study, the occurrence of MI with low, medium and high doses of a beta-blocker was compared with that with low, medium and high doses of short-acting CCBs. The higher the dose of beta-blocker the more protection against coronary events, whereas the higher the dose of CCBs the higher the risk of MI. Risk was also increased for subjects taking CCBs compared with those taking diuretic drugs. In a meta-analysis () of mortality rates in patients with ischemic heart disease randomized to receive low and high doses of nifedipine, higher doses of the short-acting calcium antagonist (>80 mg/day) increased the relative risk to more than three times that of smaller doses. There are many problems with case-control studies, and although these data must be considered as hypothesis generating, they should not be ignored.
7 Data on CCBs in Hypertension
We have limited long-term prospective data for hypertensive patients with regard to morbidity or mortality outcome with CCBs. Case-control studies suggest () an increase rather than a decrease in MI events when the short-acting CCBs are used.
In elderly hypertensive patients it has also been shown () that patients taking short-acting CCBs (diltiazem, verapamil or nifedipine) experienced an increase in the risk of CHF compared with those taking a beta-blocker or ACE inhibitor. In a recent nonrandomized, single-blind, long-term study of hypertensive subjects with slow-release nifedipine (), there was a reduction in strokes and severe arrhythmias in the treated group, but the number of CHD events was too small to reach a conclusion.
In hypertensive subjects there is suggestive evidence that the risk of an MI may actually be increased, and there is good evidence that, at least for certain CCBs, the risk of CHF is increased. These data, coupled with lack of evidence that the atherogenic process is slowed in humans as it was in animals given CCBs, seem to refute one of the major reasons advanced for using these medications in preference to others in the management of hypertension. Thus, although CCBs will lower blood pressure and, as noted, are fairly well tolerated, with the exception of edema with the dihydropyridines (short or long acting) and constipation with verapamil, there is little reason at present to suggest that these medications should be considered as preferred initial therapy.
8 Cost of CCBs—Another Issue
With the exception of a generic equivalent of verapamil slow release, the cost of CCBs, especially the newer long-acting ones, such as amlodipine and slow-release nifedipine, is considerably higher than that of other drugs, including the ACE inhibitors and A-II receptor antagonists (). Nisoldipine, a drug similar in response rate and side effects to amlodipine, is less expensive. If the CCBs possessed some unique activities or properties over and above lowering blood pressure and relieving the symptoms of angina as a result of coronary vasodilation, cost might not be a factor in considering their use; but, at least at present, these unique properties have not been identified in clinical studies.
9 Other Possible Problems With CCBs
Recent case-control studies ([33, 34]) have also suggested that the calcium antagonists increase gastrointestinal bleeding and the risk of malignancies in elderly patients. However, a large follow-up study of >11,000 patients () did not confirm an increase in the incidence of cancer when CCBs were used. One must be cautious in interpreting case-control studies; it is difficult to correct for the differences between control and actual event cases. But these confusing data are somewhat disturbing and require confirmation.
10 A-II Receptor Antagonists: Do They Offer Advantages Over Other Agents That Block the Renin–Angiotensin–Aldosterone System?
Are there any specific advantages to the newer group of compounds that block the renin–angiotensin system more peripherally at the A-II receptor sites?
At least two types of A-II receptor subtypes—AT-I and AT-II—have been identified. The AT-I receptor subtype is involved in the responses to the vasoconstrictor and aldosterone stimulation responses to A-II. Blockade of this effect results in vasodilation and a decrease in sodium and water retention (). The effect of the A-II receptor antagonist losartan on blood pressure has been compared with that of a beta-blocker (atenolol), CCBs () and ACE inhibitors (). Blood pressure-lowering effects are essentially equivalent. Unlike the action of ACE inhibitors, blockade of the renin–angiotensin system at the A-II receptor site does not interfere with the bradykinin system (Fig. 1). The buildup of bradykinin and certain prostaglandins that occurs as a result of ACE inhibition on kinase does not occur. Bradykinin is probably the major reason for a hacking, nonproductive cough that occurs in ∼15% to 20% percent of patients taking ACE inhibitors (). A-II receptor blockade does not result in a cough and represents one of the advantages of treatment with these newer agents.
In addition, the A-II receptor agents act as competitive antagonists in relation to certain other enzyme systems, such as the chymase system; A-II generation is not totally dependent on the renin–angiotensin cascade. This blockade of other pathways may result in a more complete and continuous blockade of A-II effect. This group of drugs is well tolerated, and blood pressure lowering to goal levels is achieved in ∼40% of patients.
The addition of small doses of a diuretic drug to the A-II receptor antagonist losartan adds considerably to the antihypertensive effect of this agent ([41–43]). In combination with 12.5 mg of hydrochlorothiazide, the response rate increases to >60% to 70%, with no increase in side effects. There is some diminution in the elevation of uric acid and less of a decrease in serum potassium when losartan is added to a diuretic drug. Long-term morbidity and mortality data are not available for this combination.
11 Combination Therapy
There is increasing evidence that the use of combinations of two different classes of drugs in low doses has advantages over more conventional monotherapy ([44, 45]). Cost is not increased; side effects may actually be decreased because large doses of one drug are not used; and rates of response are increased. Visits for titration are less frequent, and patient adherence is greater. In addition, taking 1 or 2 pills/day may provide some psychological benefit compared with taking 2 or 4 pills/day.
The ACE inhibitors and A-II receptor antagonists are effective antihypertensive drugs that are well tolerated by the majority of patients. Use of the AT-I receptor antagonists eliminates one of the major deterrents to the use of ACE inhibitors that block angiotensin generation as well as the degradation of bradykinin. Cough is not a side effect when these newer agents are used. In combination with a diuretic drug, an A-II antagonist reduces blood pressure to goal levels in a high percent of patients. The use of ACE inhibitors has resulted in a reduction in CHF, MI and mortality in patients with impaired systolic function and decreased ejection fraction and in post-MI patients. Data on the long-term effects of the A-II receptor antagonists are limited. Patients with diabetic nephropathy will also benefit from the use of these agents. Despite the fact that there are no data thus far on long-term morbidity and mortality in hypertensive subjects receiving medications that block the renin–angiotensin system, other than those with impaired LV function or diabetic nephropathy, there are some data to suggest that when studies presently underway are finished, such medications will be shown to be beneficial in other hypertensive subjects.
The case is not clear for the CCBs, which may lower blood pressure and decrease angina and are reasonably well tolerated by most people. There is no evidence that CHD mortality and morbidity in hypertensive subjects are reduced in long-term studies. There is some evidence that strokes are reduced with the use of a slow-release dihydropyridine formulation, but confirmatory studies are needed. There is some evidence that ischemic heart disease events may actually be increased in hypertensive patients taking CCBs compared with other agents, such as ACE inhibitors, diuretic drugs and beta-blockers. In addition, some caution must be observed in view of data suggesting an increase in event rates in subjects with ischemic heart disease receiving the rapid-release or short-acting CCBs and a possible increase in gastrointestinal hemorrhages and malignancies. Further studies are necessary to prove or disprove these initial impressions. There have not been any long-term randomized, controlled studies to date in hypertensive subjects to evaluate the effect of the longer acting formulations of these medications on morbidity and mortality.
Thus, there is some evidence that medications that interfere with the renin–angiotensin system may reduce morbidity and mortality, at least in certain subsets of hypertensive patients. In contrast, data regarding at least the shorter acting formulations of the CCBs do not indicate benefit and suggest possible harm. Preliminary data suggest a reduction in strokes in subjects treated with a long-acting or slow-release dihydropyrine ().
☆ Research support was provided by the Westchester County Hypertension Foundation, White Plains, New York and Merck & Co., Inc., West Point, Pennsylvania.
- angiotensin-converting enzyme
- angiotensin II
- AT-I, AT-II
- angiotensin II receptor subtypes I and II
- calcium channel blocking agents
- coronary heart disease
- congestive heart failure
- JNC V
- Fifth Joint National Committee on Detection, Evaluation and Treatment of High Blood Pressure
- left ventricular
- left ventricular hypertrophy
- myocardial infarction
- Received November 7, 1996.
- Revision received February 13, 1997.
- Accepted February 26, 1997.
- The American College of Cardiology
- MRC Working Party
- Parving HH,
- Andersen A,
- Smidt U,
- Christiansen JS,
- Oxenhoff B,
- Svednsen PA
- Parving HH,
- Andersen AR,
- Smidt UM,
- et al.
- Erley CM,
- Haefele U,
- Heyne N,
- et al.
- Gansevoort RT,
- de Zeeuw D,
- Shahinfar S,
- Redfield A,
- de Jong PE
- Crozier I,
- Ikram H,
- Awan N,
- Cieland J,
- et al.
- Kostis JB
- Waters D,
- Lesperance J,
- Francetich M,
- et al.
- Furberg CD,
- Psaty BM,
- Myers JV
- Pahor M,
- Guralnik JM,
- Salive ME,
- Corti MC,
- Carbonin P,
- Havlik RJ
- on behalf of the Bezafibrate Infarction Prevention Study participants,
- Braun S,
- Boyko V,
- Behar S,
- Reicher-Reiss H,
- et al.
- Dahlof B,
- Keller SE,
- Makris L,
- Goldberg AI,
- Sweet CS,
- Lim NY
- Moser M
- Held PH,
- Yusuf S,
- Furberg CD
- 1 The Case for ACE Inhibitors
- 2 ACE Inhibitors in Diabetes and Renal Disease
- 3 A-II and Aldosterone Inhibition in Heart Failure
- 4 The Case for CCBs
- 5 Effects of CCBs in Patients With Ischemic Heart Disease
- 6 Effects of CCBs on Progression of Atherosclerosis
- 7 Data on CCBs in Hypertension
- 8 Cost of CCBs—Another Issue
- 9 Other Possible Problems With CCBs
- 10 A-II Receptor Antagonists: Do They Offer Advantages Over Other Agents That Block the Renin–Angiotensin–Aldosterone System?
- 11 Combination Therapy
- 12 Summary