The b-blockers were among the first therapeutic interventions designed to limit the size of an AMI. In most of these studies, all patients with AMI were included together regardless of the direction of the ST-segment deviation on admission. Norris and colleagues in New Zealand (11,12) demonstrated in 1978 that early administration of b-blockers decreased the size of AMI measured enzymatically (as a function of creatine-kinase
Fig. 1. Results of 51 randomized clinical trials of the effects of oral or iv followed by oral P-blockers for the treatment of AMI (odds ratios and 95% CI for effect of treatment on mortality). Reproduced with permission from ref. 60.
enzyme release) or by reduction of ST-segment elevation. Some of the early studies of P-blockers to reduce infarct size (13) were limited by the lack of appreciation that the window of time within which myocardium may be salvaged was in the range of 6-12 h. In the Multicenter Investigation of the Limitation of Infarct Size (MILIS) study, for example, P-blockers were administered up to 24 h after onset of chest pain (13).
More recent studies using a more appropriate design to administer the P-blocker within an appropriate time window have definitively demonstrated a benefit associated with P-blocker therapy (Fig. 1) (14,15). In the largest trial, the First International Study of Infarct Survival (ISIS-1) (16), over 16,000 patients with suspected MI were treated with immediate intravenous (iv) atenolol, 5-10 mg, within 12 h of the onset of symptoms, followed by 100 mg daily. The mortality difference between those receiving atenolol and the controls was evident by the end of d 1; the 7-d mortality was reduced
Fig. 2. Sudden death, other death, and nonfatal reinfarction in long-term P-blocker trials that reported these end points separately (odds ratios [active/control], together with approximate 95% confidence ranges). See original article for citation of specific trials. Reproduced with permission from ref. 14.
from 4.3 to 3.7% (p < 0.02). Meta-analyses from 27 randomized trials, totaling about 27,000 patients, indicate that early iv (followed by oral) b-blockers reduced mortality by 13% in the first week (95% confidence interval [CI]:-2-25;p < 0.02) (14,17). The mortality reduction was greatest in the first 2 d (about 25%), supporting the value of early initiation of b-blockade (17). Early treatment also reduced nonfatal reinfarction by about 19% (95% CI -5 to -33;p < 0.01) and nonfatal cardiac arrest by about 16% (95% CI:-2-30; p < 0.02). Composite end points of death, nonfatal reinfarction, and nonfatal arrest were reduced by 16% (p < 0.001) (17). Data from the ISIS-1 trial suggest that the reduction in mortality is largely due to prevention of cardiac rupture and ventricular fibrillation (16). Detailed analyses of the results based on various subgroups (initial heart rate, risk category, presence or absence of ventricular arrhythmia, and so on) indicated a benefit in all groups.
When b-blockers are used in conjunction with thrombolytic therapy, they provide incremental benefit, particularly if they can be administered early after the onset of infarct symptoms. In the Thrombolysis in Myocardial Infarction (TIMI)-II trial (18), patients with persisting ST-segment elevation, who were randomized to receive early metoprolol (15 mg iv, followed by oral metoprolol 50 mg twice daily (bid) for 1 d and then 100 mg bid thereafter) in addition to iv alteplase, experienced a 49% lower incidence of subsequent nonfatal reinfarction (p = 0.02) and a 27% lower incidence of
Fig. 3. Mortality in long-term P-blocker trials, by ancillary properties of agent tested (odds ratios [active/control], together with approximate 95% CI. See original article for citation of specific trials. Reproduced with permission from ref. 14.
recurrent ischemia (p = 0.005) compared to those patients randomized to receive meto-prolol only orally, beginning 6 d after the acute event. Those patients who were treated within 2 h of symptom onset had the greatest reduction of the composite end point of death or reinfarction compared to those treated to only late oral metoprolol.
A number of studies have classified the mechanism of death as sudden or nonsudden, based on the duration of time from the onset of symptoms to actual death. Sudden death is variably defined as "instantaneous" to "within 2 h of symptoms" and is presumably due to arrhythmias or cardiac rupture; nonsudden deaths are those occurring later after the onset of symptoms, presumably owing to nonarrhythmic causes, such as reinfarc-tion, and may include a few noncardiac deaths. Tabulation of the results from the available studies indicates a highly significant reduction of approx 30% in the incidence of sudden death and a nonsignificant reduction of only about 12% in the incidence of nonsudden death (Fig. 2) (14). The fact that P-blockers were particularly effective in reducing both sudden death and mortality among patients with complex ventricular ectopy at baseline (19) suggests that P-blockers exert their beneficial effect primarily by reducing the frequency and severity of arrhythmias (20).
It is striking that the long-term mortality benefits of the P-blockers following an index MI (i.e., secondary prevention) extend to most members of this class of agents (14). There does not seem to be a significant difference agents with or without cardioselectivity (Fig. 3). However, the presence of intrinsic sympathomimetic activity reduced the benefit to
Fig. 4. (A) Relation between reduction in heart rate (difference between treatment groups) and percentage of reduction in mortality in large, prospective, double-blind trials with P-blockers. Open circles, P-blockers with intrinsic sympathomimetic activity: r = 0.6:p < 0.05. See original article for citation of specific trials. Reproduced with permission from ref. 21. (B) Relation between reduction in heart rate and percentage of reduction in recurrent nonfatal infarctions in large, prospective, double-blind trials with P-blockers. Open circles, P-blockers with intrinsic sympathomimetic activity; r = 0.59;p < 0.05. See original article for citation of specific trials. Reproduced with permission from ref. 21.
nonsignificance (odds ratio 0.90; 95% CI: 0.77-1.05) (Fig. 3) (14). Reduction in heart rate appears to be a critical feature associated with the protective effect of P-blockers. Indeed, there is a significant relationship between the magnitude of heart rate reduction observed on the active agent and the magnitude of reduction in mortality (Fig. 4A) (21).
Many of the large-scale clinical trials have also reported the effects of long-term P-blocker use on nonfatal reinfarction. Results from pooled analyses indicate that P-
blocker use is associated with an odds ratio of 0.74 (95% CI: 0.66-0.83;p < 0.001). As observed for mortality, there is also a significant relationship between the magnitude of reduction in heart rate and the reduction in nonfatal recurrent MI (r = 0.54; p < 0.05) (Fig. 4B) (21). This observed benefit of reducing nonfatal reinfarction is in addition to the benefit on mortality.
The magnitude of benefit from long-term use of a b-blocker is also dependent on the patient's risk of mortality associated with their index MI (Table 1). Post hoc analyses of data from the Beta Blocker Heart Attack Trial (BHAt) (22) indicate that those MI patients without electrical or mechanical complications experienced only a 6% relative benefit from the use of propranolol. MI patients with electrical complications experienced a 52% relative benefit, those with mechanical complications experienced a 38% relative benefit, and those with both mechanical and electrical complications experienced a 25% relative benefit. Considering the low cost of routine b-blocker use and its substantial benefit, such therapy has a relatively favorable cost-effectiveness ratio: an estimated cost of therapy per year of life saved would be $13,000 in low-risk patients, $3600 in medium-risk patients, and $2400 in high-risk patients (23).
The benefits from routine b-blocker use seem to persist as long as the active agent is continued (24-26). It is, therefore, most appropriate after MI to maintain b-blocker therapy indefinitely in patients who can tolerate it (27). The benefits of a b-blocker in long-term secondary prevention appear to extend to most patient subgroups. The Beta-Blocker Pooling Project (28) combined the results of nine large trials and found that although high-risk patients were most likely to benefit from b-blocker therapy, lower-risk patients also benefited, even though the absolute and relative benefits were small. The experience using b-blockers in the elderly is limited, but available data indicate that the benefit may even be greater in patients older than 50-60 yr than in younger patients. Benefit appeared to be similar in both men and women.
The recently published CAPRICORN trial (29) was a multicenter double-blind randomized controlled trial of 1959 patients with a definite MI 3-21 d prior to enrollment with a left ventricular (LV) ejection fraction of 40% or less, who were receiving concurrent treatment with an angiotensin-converting enzyme (ACE)-inhibitor. Patients were randomly assigned to receive carvedilol or placebo with an upward titration of the dose to a maximum of 25 mg bid. The primary end point was all-cause mortality and hospital admission. After a mean follow-up of 1.3 yr, all-cause mortality was lower in the carvedilol than in the placebo group (12 vs 15%, p = 0.03), but there was no significant difference in the co-primary end points of death and hospitalization. In addition, the patients treated with the b-blocker had lower rates of cardiovascular mortality (11 vs 14%, p = 0.024) and nonfatal myocardial infarctions (3 vs 6%, p = 0.014). The benefits of long-term therapy with carvedilol in post-MI patients complicated by LV dysfunction were in addition to the effects of ACE-inhibitors, which were prescribed in up to 98% of the enrolled patients.
Given the beneficial results of other b-blocker trials in the post-MI setting, the CAPRICORN trial might be generalized to other b-blockers in providing yet another strong evidence that, unless contraindicated, all post-MI patients should be maintained on long-term b-blocker therapy in addition to other proven therapies, such as ACE inhibitors. The side effects from prolonged b-blocker use have generally been minor and are similar to those seen with placebo (30). In studies that report it, the incidence of heart failure is slightly but significantly higher in patients receiving b-blocker (5.9%) than in
patients receiving placebo (5.4%) (pooled odds ratio 1.16; 95% CI: 1.01-1.34) (14). However, even patients with a history of mild or moderate congestive heart failure (CHF) actually experienced greater benefit from b-blockade than did patients without that condition (20).
Was this article helpful?