| Norman M Kaplan, MD Burton D Rose, MD |
Apr 3, 2000 |
Calcium channel blockers are widely used in the treatment of hypertension, angina pectoris, cardiac arrhythmias, and other disorders and the longer-acting preparations have been prescribed with increasing frequency since 1989 (show figure 1). This card will review the major side effects associated with these agents and the current controversy concerning their effect on coronary events, mortality, gastrointestinal bleeding, and the development of cancer.
TYPES OF CALCIUM CHANNEL BLOCKERS ！ The calcium channel blockers currently available are divided into two major categories based upon their predominant physiologic effects: the dihydropyridines which preferentially block the L-type calcium channels; and verapamil and diltiazem [1,2,3]. The L-type calcium channels are responsible for myocardial contractility and vascular smooth muscle contractility; they also affect conducting and pacemaker cells.
Dihydropyridines ！ The dihydropyridines are potent vasodilators that have little or no negative effect upon cardiac contractility or conduction. They can be further divided into three categories based upon half-life and effect on contractility:
• Short-acting liquid nifedipine
• Longer-acting formulations with little cardiac depressant activity ！ felodipine, isradipine, nicardipine, nifedipine GITS and CC, and nisoldipine
• Long-acting agents with no cardiac depressant activity ！ amlodipine, lacidipine
Verapamil and diltiazem ！ Verapamil and, to a lesser extent, diltiazem are less potent vasodilators but have negative effects upon cardiac conduction and contractility. (See "Calcium channel blockers in the treatment of cardiac arrhythmias").
KNOWN SIDE EFFECTS ！ The side effects that may be seen with the calcium channel blockers vary with the agent that is used. The potent vasodilators can, in 10 to 20 percent of patients, lead to one or more of the following: headache, dizziness or lightheadedness, flushing, and peripheral edema [1]. The peripheral edema, which is infrequent with verapamil [4], is related to redistribution of fluid from the vascular space into the interstitium, possibly induced by vasodilation which allows more of the systemic pressure to be transmitted to the capillary circulation [5,6]. In one study of 12 healthy subjects, for example, a single dose of nifedipine increased the foot volume despite also increasing sodium excretion [6]. Thus, treatment of this form of edema with a diuretic will not relieve the edema. On the other hand, edema is much less common when a dihydropyridine is given with an angiotensin converting enzyme inhibitor [7]. This effect is probably related to venodilation by the ACE inhibitor which helps remove the fluid sequestered in the capillary bed by the arteriolar dilation from the calcium channel blocker. This form of combination therapy is likely to become much more common since the Food and Drug Administration has approved fixed (low) dose combination preparations of these drugs.
The major adverse effect with verapamil is constipation, which can occur in over 25 percent of patients [1].
Effects on cardiac function ！ Verapamil and, to a lesser degree, diltiazem can diminish cardiac contractility and slow cardiac conduction [2]. As a result, these drugs are relatively contraindicated in patients who are taking beta blockers or who have severe left ventricular systolic dysfunction, sick sinus syndrome, and second or third degree atrioventricular block.
The dihydropyridines have less cardiac depressant activity in vivo for two reasons:
• The doses employed are limited by the peripheral vasodilation; as a result, plasma levels sufficient to impair contractility and atrioventricular conduction are not achieved [8]
• Acute vasodilation leads to a reflex increase in sympathetic activity that can counteract the direct effect of calcium channel blockade [9].
Use in congestive heart failure ！ The negative inotropic effects of the calcium channel blockers and perhaps the increased sympathetic activity induced by the dihydropyridines have made these agents less useful in the treatment of patients with congestive heart failure due to impaired systolic function. In this setting, the administration of a calcium channel blocker may be associated with an increased risk of cardiac decompensation [10].
Different findings have been reported with amlodipine, which does not induce sympathetic stimulation. The Prospective Randomized Amlodipine Survival Evaluation (PRAISE) trial found no overall benefit when amlodipine was added to ACE inhibitors and diuretics. However, there was a significant 45 percent decrease in mortality in patients with class III and IV heart failure caused by dilated nonischemic cardiomyopathy [11]. (See "Calcium channel blockers in chronic congestive heart failure").
Possible mechanisms of increased cardiovascular mortality ！ With respect to a possible increase in cardiovascular mortality described in the next section, there are several potential mechanisms whereby calcium channel blockers may have a deleterious effect [12,13]:
• Enhanced sympathetic activity resulting from hypotension can result in enhanced activity of normal and ectopic pacemaker tissue, an increase in heart rate, and elevation in serum norepinephrine concentrations. These changes are primarily observed with the short-acting, but not long-acting dihydropyridines (show table 1) [14].
• Proarrhythmia resulting from interference with normal sinus and atrioventricular nodal function with the development of impulse conduction abnormalities, including transient asystole. Another possible mechanism is potential myocardial calcium overload resulting in afterdepolarizations and triggered activity (See "Mechanisms and treatment of proarrhythmia").
• Negative inotropic activity
• Proischemic effects (from coronary steal)
SERIOUS ADVERSE EFFECTS ！ Several recent studies have raised concern about the possible long-term safety of therapy with calcium channel blockers. These issues need to be reviewed in detail in order to make a reasoned decision about the optimal use of these drugs.
Increased mortality after an acute myocardial infarction ！ Large doses of short-acting nifedipine may increase mortality in patients in the immediate post-myocardial infarction period (show figure 2) [15]. Although the strength of this association appears to have been exaggerated by the inclusion of inappropriate data [16], the potential for harm from the profound hypotension and sympathetic activation induced by such large doses of short-acting nifedipine is incontrovertible in these hemodynamically vulnerable patients (show table 1).
This danger from short-acting agents may not apply to patients who have survived 12 months or longer after a myocardial infarction, even if they continue to have angina [17]. Despite this observation, there is no compelling reason to use short-acting calcium channel blockers in such patients.
The effect of verapamil and diltiazem in the post-MI patient is discussed in detail elsewhere. (See "Calcium channel blockers in acute myocardial infarction"). It is useful, however, to review the results of two of the major trials:
• The DAVIT-II study results suggest that a subset of patients may benefit from verapamil. Almost 1800 patients were randomized to verapamil or placebo [18]. After a mean follow-up of 16 months, there was a 21 percent reduction in deaths in patients receiving verapamil (95 versus 119 deaths in the placebo group), a difference that did not reach statistical significance. The observed benefit with verapamil was confined to patients without a history of congestive heart failure.
• The Multicenter Diltiazem Post Infarction Trial Research Group study is the only major clinical trial which has examined the efficacy of diltiazem [19]. Diltiazem was found to exert no significant effect on mortality, which was 13.5 percent with both diltiazem and placebo. There was, however, a nonsignificant trend toward a benefit in patients with no evidence of pulmonary congestion on chest x-ray, while there was an excess of deaths in the group with pulmonary congestion.
It is therefore possible that calcium channel blockers may benefit a few selected patients after MI, particularly those with a non-Q wave infarct. However, the level of benefit, if present, is small. There are also subsets of patients in whom certain calcium channel blockers may increase mortality, such as those with congestive heart failure. (See "Calcium channel blockers in chronic congestive heart failure").
Increased risk for an acute myocardial infarction in hypertensive patients ！ A 1995 report which suggested that calcium channel blockers increased the risk for an acute myocardial infarction received a great deal of publicity in the lay press [20]. This case-control study compared 623 hypertensive patients who had had a myocardial infarction to 2032 patients who had not. Patients treated with a short-acting calcium channel blocker (nifedipine, diltiazem, or verapamil) had a relative risk of 1.6 for myocardial infarction, a finding that was not seen with diuretics, beta blockers, or angiotensin converting enzyme inhibitors. The relative risk was increased (2.88 to 3.33) only in patients treated with high doses; the risk with low doses of a calcium channel blocker was similar to that seen with other antihypertensive drugs. "High" and "low" referred to daily doses greater than or less than 30 mg of nifedipine, 180 mg of diltiazem, and 240 mg of verapamil (show figure 3).
The Nurses' Health Study of 14,617 women with treated hypertension reported similar results [21]. Among those women who used calcium channel blockers as monotherapy, the relative risk of myocardial infarction after adjusting for age and baseline prevalence of ischemic heart disease and other cardiovascular risk factors was 1.64. When the use of any calcium channel blocker, as monotherapy or in combination, was compared to that of any other antihypertensive drug, the risk ratio was 1.42. This association was seen in smokers, but was not seen in nonsmokers.
These provocative observational findings are, by nature, uncontrolled and inconclusive. Many confounding variables could be responsible, with two being of particular concern:
• Antihypertensive therapy was selected by the patients' physicians. It is therefore probable that patients receiving a calcium channel blocker had more underlying coronary disease. The short-acting calcium channel blockers that were used were approved for the treatment of angina and not for hypertension, were more expensive, and more difficult for patients to take on a three times a day schedule than the approved drugs that were available; thus, they were likely used more in hypertensives with underlying coronary disease.
• The calcium channel blockers used were short-acting verapamil, diltiazem, and nifedipine preparations. Currently used long-acting calcium channel blockers were not included. Studies in hypertensive patients in general have not found evidence of increased coronary risk from long-acting calcium channel blockers which, as noted above, do not share the possible harmful cardiovascular effects of the short-acting preparations (show table 1) [22,23,24,25,26]. A recent case-control study evaluated 189 hypertensive patients who had a first cardiovascular event and compared them to 189 matched controls [24]. When compared to therapy with beta blockers, the odds ratio for a cardiovascular event was significantly increased with short-acting calcium channel blockers (OR 3.88 compared to beta blockers) but was not increased with long-acting preparations (OR 0.76). Even more assuring are the data from prospective, controlled studies in which long-acting calcium channel blockers have been shown to reduce the incidence of coronary events [25,26].
However, recent trials in hypertensive diabetics suggest that even the long-acting dihydropyridines may not provide protection against coronary disease as seen with ACE inhibitors. The Appropriate Blood Pressure Control in Diabetes (ABCD) compared first-line therapy with enalapril or nisoldipine, each in 235 patients, for five years [27]. Patients treated with enalapril had fewer fatal and nonfatal myocardial infarctions (4 versus 24, risk ratio 7.0). Similar differences were noted in the FACET trial in which 380 hypertensive type 2 diabetics were treated with fosinopril, amlodipine, or both for three years [28]. A cardiovascular endpoint occurred in 27 of the 141 patients on amlodipine, 14 of the 131 on fosinopril, but in only 4 of the 108 treated with both drugs.
The incidence of myocardial infarctions in patients on nisoldipine in the ABCD trial was similar to that seen in nontreated hypertensive diabetics [27]. As a result, the authors interpreted these data as showing a benefit from the ACE inhibitor but no increased risk from nisoldipine.
This conclusion of lack of an adverse effect of calcium channel blockers in diabetics is supported by results from the Hypertension Optimal Treatment (HOT) trial in which hypertensive patients were randomized to target diastolic pressures of
90,
85
mmHg, or 80
mmHg; the mean diastolic blood pressures attained for these three groups were
85.2, 83.2, and 81.1 mmHg, respectively [29].
Initial therapy was with felodipine
with other drugs added as necessary. Among the subset of 1501 diabetic
patients, the relative risk of a cardiovascular event was significantly
reduced in the 80
mmHg group compared to the 90
mmHg group (relative risk = 0.49).The major conclusion from these trials is that an ACE inhibitor should be the initial choice of therapy for all hypertensive diabetics, often with a diuretic. If the blood pressure is not controlled to below 130/80, a long-acting dihydropyridine may be used [30].
Increased MI and mortality in congestive heart failure ！ There are conflicting data on the effect of calcium channel blockers on myocardial infarction and mortality in patients with left ventricular dysfunction. A retrospective analysis evaluated the outcome in 6797 patients entered into the SOLVD trials, one-third of whom were taking a short-acting calcium channel blocker, mostly diltiazem [31]. There was an increased incidence of myocardial infarction among patients treated with a calcium channel blocker in those randomized to enalapril (11.5 versus 7.5 percent) and those receiving placebo (14.4 versus 9.3 percent). The adjusted risk ratio for all-cause mortality associated with calcium channel blocker use was 1.14 (p = 0.045). In contrast, data from the SAVE trial, which randomized 2231 patients with an acute myocardial infarction and a left ventricular ejection fraction
40
percent to captopril
or placebo, found no adverse effect of concurrent calcium channel blocker use
(which was present in over 40 percent of patients) [32].
Calcium channel blockers did not alter the benefit of captopril.Increased mortality, gastrointestinal bleeding, and cancer in the elderly ！ Four studies from the same investigators reported the mortality and morbidity of approximately 1000 hypertensive patients among a total of 8000 elderly individuals screened as part of a three community epidemiologic study. Among the hypertensive patients, approximately 200 individuals (20 percent) were taking calcium channel blockers.
• The first study reported an increased mortality rate for patients in whom a short-acting nifedipine was being administered in 1988 and whose status was derived from Medicare records in 1992 [33].
• The second described an increased rate of gastrointestinal hemorrhage in those taking short-acting verapamil or diltiazem, but not nifedipine [34].
• The third described an increase in various cancers in those taking short-acting verapamil or nifedipine, but not diltiazem [35].
• The fourth evaluated the incidence of cancer among a larger portion of the same population including 450 subjects who were taking calcium channel blockers [36]. A multivariate analysis of possible confounding variables was used to compare the two nonrandomized populations. The following results were obtained:
• An increased hazard ratio of 1.72 for developing cancer was found for patients taking calcium channel blockers
• The increased risk for cancer was dose-dependent and was found for a variety of different types of cancers
• No increased risk was uncovered among those taking cardiovascular medications other than a calcium channel blocker
These cohort studies on the same population of elderly people (average age 78, range from 71 to 96) have multiple potential faults:
• In all four studies, no attempt was made to determine whether patients who were reportedly taking calcium channel blockers at the one survey point in 1988 had actually continued to take these drugs. As a result, a patient considered among those who were taking a calcium channel blocker may have only received the drug for one day or not at all.
• The total number of reported adverse effects was small since only a few patients were taking different antihypertensive drugs.
• The much larger population (over 4000 patients) not taking a beta blocker, ACE inhibitor, or calcium channel blocker were excluded. About one-half of these patients were taking a diuretic.
• Patients on calcium channel blockers were more likely to be hospitalized since those given calcium channel blockers were in poorer health than those not receiving these drugs. As a result, the presence of gastrointestinal hemorrhage or cancer was more likely to be uncovered. The same including bias wherein sicker patients are given certain drugs which are then blamed for worse outcomes could also explain a description of more severe white matter changes on MRI and worse performance on mental examination in hypertensive patients treated with calcium channel blockers or loop diuretics [37].
Because of the multiple problems inherent in uncontrolled, retrospective case-control and cohort observational studies, conclusions drawn from these reports are frequently found to be in error after controlled, prospective, randomized studies are performed [38]. Furthermore, the authors of these papers have extended the putative associations found from the use of short-acting calcium channel blockers to that of long-acting formulations. This conclusion is not warranted since the two groups of calcium channel blockers greatly differ in their effects (show table 1) [9].
With respect to the claim of enhanced cancer risk, three larger case-control studies found no increase in the incidence of malignancy when patients treated with calcium channel blockers were compared to those who were not [39,40,41]. Similar findings have been noted in two large, prospective, randomized hypertension treatment trials. A lower incidence of cancer was found among the one-half of the 4695 patients in the Syst-Eur trial who were treated with the calcium channel blocker nitrendipine [26]. In addition, there may be an association between the diagnosis of hypertension and renal cell carcinoma that is independent of antihypertensive medications, suggesting a confounding effect in the previous reports claiming a relationship between calcium channel blockers and malignancy [42].
Lack of an increase in noncardiovascular events was also noted in the HOT trial in patients treated with felodipine [29]. In addition, no enhanced incidence of gastrointestinal bleeding was found in those using a calcium channel blocker in a large Canadian study of nearly 35,000 patients [43].
Summary ！ In summary, these studies point out the potential risk of short-acting calcium channel blockers [24]. However, several additional reports have been published which show no increased risk from the longer-acting calcium channel blockers in patients who are postmyocardial infarction, or who have congestive heart failure or hypertension [11,25,26,44,45]. As a result, a prudent recommendation is that the use of short-acting nifedipine be curtailed [46]. In addition to the apparent increase in long-term risk, the acute use of sublingual nifedipine in patients with severe hypertension can also be detrimental, possibly inducing hypotension which can lead to cerebrovascular or coronary ischemia [47]. (See "Severe asymptomatic hypertension (hypertensive urgencies)").
It is our feeling, and the opinion of most experts who have reviewed the evidence [48], that the above findings should not deter physicians or patients from using long-acting calcium channel blockers for the treatment of hypertension and angina when they are clearly indicated. These include elderly patients with isolated systolic hypertension and patients with one of the following coexisting conditions:
• Angina pectoris
• Raynaud's phenomenon
• Renal transplantation (to counteract cyclosporine-induced renal vasoconstriction)
• Asthma or chronic obstructive pulmonary disease
• Failure to respond to or tolerate other medications
It should be appreciated, however, that there is usually no compelling reason to use a calcium channel blocker as a first-line agent in the treatment of most patients with essential hypertension. In addition to increased cost, the calcium channel blockers lack the cardiovascular benefits of some other antihypertensive agents in certain clinical settings. As examples, beta blockers and perhaps ACE inhibitors are beneficial after acute myocardial infarction; ACE inhibitors and, in many cases, beta blockers are beneficial in patients with congestive heart failure; and ACE inhibitors are beneficial in patients with diabetic nephropathy and certain other forms of chronic renal failure. (See appropriate cards).
The potential importance of these differences was suggested in the STOP hypertension-2 trial which compared the efficacy of older (diuretics and beta blockers) and newer (ACE inhibitors and calcium channel blockers) antihypertensive drugs in elderly subjects [49]. Although the degree of blood pressure control (194/98 at baseline to 159/81) and, at 4.5 year follow-up, the combined endpoint of fatal and nonfatal stroke or myocardial infarction and other cardiovascular mortality were the same in all groups, the frequency of myocardial infarction and heart failure was significantly lower in patients receiving ACE inhibitors compared to those receiving calcium channel blockers (felodipine or isradipine.
CALCIUM CHANNEL BLOCKER OVERDOSE ！ Patients ingesting more than five to ten times the usual dose can develop signs of severe intoxication including drowsiness, confusion, hyperglycemia (due to decreased insulin release), and most importantly cardiovascular collapse with hypotension and metabolic acidosis [50]. Patients with preexisting sinus node or conducting system disease may develop sinus bradycardia or second or third degree heart block. The symptoms and prognosis are worse with verapamil than diltiazem, and with sustained release preparations.
The general approach to any poisoned patient must include the following elements:
• Evaluation, including the recognition that poisoning has occurred, identification of the agents involved, assessment of severity, and prediction of toxicity. (See "General approach to drug intoxications").
• Management, consisting of supportive care, prevention of drug absorption, and, when appropriate, the administration of antidotes and enhancement of drug elimination. (See "Decontamination of poisoned patients" and see "Enhanced elimination of poisons").
The calcium channel blockers are poorly removed by hemodialysis or hemoperfusion, since they are highly protein bound and have extensive tissue distribution. Treatment consists of fluid replacement and the administration of intravenous calcium. Three different regimens have been used [50]:
• 10 percent calcium chloride, 0.2 mL/kg up to a maximum of 10 mL infused over five minutes. The dose can be repeated every 15 to 20 minutes up to four times, if necessary.
• 10 percent calcium chloride, 0.2 mL/kg to a maximum of 10 mL given as a continuous infusion per hour.
• 10 percent calcium gluconate, up to a maximum of 20 to 30 mL infused over 10 five minutes. The dose can be repeated every 15 to 20 minutes up to four times, if necessary.
Pressor agents are given as required. Small case series have suggested a beneficial effect of an insulin-dextrose infusion, and larger trials of this therapy are planned [51].