The Most Important Issue: Use Beta Blockers

Despite all of the currently known beneficial effects of β blockers, questions remain about their use. This issue of Congestive Heart Failure includes three excellent papers detailing various aspects of using β blockers for heart failure. Chavey et al.1 concisely and expertly review some of the unresolved issues on the use of β blockers in heart failure, including a comparison of carvedilol and metoprolol and an analysis of concomitant therapies (angiotensin-converting enzyme, angiotensin receptor blocker, spironolactone). Maack et al.2 extensively compare mechanisms of carvedilol and metoprolol, the two US Food and Drug Administration-approved β blockers for heart failure in this country. The outcome comparisons, however, are limited because there are no direct comparisons of carvedilol and extended-release metoprolol succinate (and no direct comparative trials currently in progress or being planned). Finally, Abraham3 presents algorithms for switching from metoprolol to carvedilol or vice versa for perceived potential differential benefits between the two β blockers. It is clear that there is one question of enduring debate: Is there a greater benefit to selective β-1 blockade (metoprolol or bisoprolol), or do the ancillary properties of the β nonselective carvedilol (β-1 and β-2 antagonism) with vasodilating properties (α-antagonism) and in vitro antioxidant effects4,5 provide specific benefits above and beyond β-1 blockade itself? Unlike β blockers, α blockers have not shown sustained benefits in chronic heart failure. Despite acute hemodynamic benefits, patients treated with these drugs develop hemodynamic tolerance within days of continued therapy.6 Furthermore, in the Vasodilator Heart Failure Trial I (V-HeFT-1) study, there was no survival benefit of prazosin vs. placebo.7 Perhaps α-blocking effects might persist in the presence of β blockade during chronic use.8 In such a study, we compared chronic metoprolol therapy to combination therapy with metoprolol and the α blocker doxazosin. After 3 months of chronic therapy there was no hemodynamic difference between β blockade and combined αβ blockade, implying tolerance to α blockade despite the presence of β blockade. In the ALLHAT9 hypertension trial, the α- blockade arm was prematurely terminated due to an increased incidence of heart failure and also cardiac death in those hypertensive patients taking doxazosin. Kubo and colleagues10 compared the chronic effects of carvedilol and metoprolol on calf hemodynamics. No difference was seen between metoprolol and carvedilol, implying tolerance to the hemodynamic α effects of carvedilol. Thus, there are no demonstrated benefits of a blockade in treating chronic heart failure. Gilbert et al.11 indirectly compared metoprolol and carvedilol by combining two different studies, each of which was similar in design (but not identical). They showed that β receptors are up-regulated with metoprolol tartrate while there is no change in these receptors with carvedilol. The clinical significance of this is unknown. An aspect of the debate on efficacy of different β blockers involves oxidative stress; does this play a significant role in the pathophysiology of heart failure, and are specific pharmacologic agents effective due to their antioxidant effects? There is evidence that free radicals play an integral role in the pathogenesis and progression of heart failure.12–15 Belch et al.13 showed that chronic heart failure patients have evidence of increased oxidative stress compared with normal controls, and more recent studies showed that markers of oxidative stress increase with worsening New York Heart Association (NYHA) functional class.16,17 Furthermore, there is some evidence that antioxidant therapies may be beneficial in improving outcomes of chronic heart failure patients.18 Our group19 randomized 67 patients with chronic heart failure, predominantly NYHA class III (72%), to receive either carvedilol or metoprolol tartrate in addition to background medications. Symptom scores, exercise parameters, and ejection fractions improved in both groups, with no significant between-group differences detected. Interestingly, thoibarbituric acid reactive substances—a marker of lipid peroxidation and oxidative stress—decreased significantly in response to treatment with both metoprolol and carvedilol, with no between-group difference (Figure 1). This occurred despite the fact that carvedilol has potent antioxidant and free radical scavenging activity in vitro and metoprolol has no known antioxidant properties. Instead, the decrease in thoibarbituric acid reactive substances appeared to correlate with improving heart failure status. Thus, oxidative stress may be a marker that characterizes heart failure and may be beneficially effected by appropriate β-blocker therapy without a specific carvedilol effect. Thoibarbituric acid reactive substances (TBARS) values for those who completed the protocol at baseline, month 4, and month 6 for metoprolol and carvedilol, respectively. Values are mean±standard error of the mean; p values inside box are within-group paired t tests between baseline and month 6; p value above box is between groups by ANOVA for all time points. Printed with permission from Circulation. 1999;99:2645–2651. Similarly, Sanderson et al.20 from Hong Kong reported a double, blind-positive-controlled comparison (51 patients) showing both metoprolol tartrate and carvedilol therapies were associated with a highly significant improvement in symptoms, exercise capacity, and left ventricular ejection fraction with no significant differences between the two β blockers. Metra et al.,21 in a double-blind study, compared carvedilol and metoprolol tartrate with hemodynamic end points. They reported similar parallel improvements with both drugs. There was a greater improvement in ejection fraction with carvedilol compared with metoprolol while there was a greater improvement in exercise tolerance with metoprolol. It is conceivable that the upregulation of recruitable β receptors with metoprolol compared with carvedilol reported by Gilbert et al.11 may explain the divergent exercise and hemodynamic responses. Finally, even the benefit of β-2 antagonism has recently been called into question. In an elegant experiment with adult rat myocytes, Communal et al.22 showed that β-2 antagonism actually caused an increase in apoptosis while β-1 antagonism decreased apoptosis. Three appropriately powered placebo-controlled β blocker mortality trials in heart failure—Cardiac Insufficiency Bisoprolol Study II (CIBIS-II),23 Carvedilol Prospective Randomized Cumulative Survival Study (COPERNICUS),24 and Metoprolol CR/XL Randomised Intervention Trial in Congestive Heart Failure (MERIT-HF)25—using bisoprolol, carvedilol, and metoprolol succinate, respectively, showed a remarkable concordance in the primary end point of all-cause mortality with a risk reduction of 34%–35%. The recently published Carvedilol or Metoprolol European Trial (COMET) study26 directly compared metoprolol tartrate and carvedilol with a mortality end point. In this trial, 3029 NYHA class II–IV patients (ejection fraction ≤35%) were randomized to receive carvedilol 25 mg b.i.d. or metoprolol tartrate 50 mg b.i.d. in addition to their background therapy. The groups were balanced at baseline with an average age of 62 years, 80% male, ejection fraction 26%, 98% on background angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, and 96% classified as NYHA class II or III. After a mean study period of nearly 5 years, there were 512 deaths in the carvedilol group and 600 deaths in the metoprolol tartrate group. This 17% reduction in mortality favoring carvedilol had a p value of 0.0017. The coprimary combined end point of death or hospitalization was not significantly different between groups. It should be emphasized that the metoprolol studied in COMET was the tartrate preparation, not the succinate once-daily formulation used in MERIT-HF. Additionally, the target dose of metoprolol tartrate was 50 mg b.i.d. in contrast to the 200 mg target dose of metoprolol succinate in MERIT-HF. Does this resolve the β-1 vs. β-1–β-2 blockade issue? Dargie,27 in an accompanying editorial, addresses the crucial issues of dosing and the comparability of metoprolol tartrate (used in COMET) with extended-release metoprolol succinate (used in MERIT-HF). He concludes that the definitive comparison of carvedilol and metoprolol succinate at doses proven to reduce mortality remains unanswered. One crucial unanswered question is the dosing of β blockers. In the COPERNICUS trial,24 the target dose of carvedilol was 25 mg b.i.d, and in the US Carvedilol trials,28 patients weighing 85 kg or more were dosed at 50 mg b.i.d. In the MERIT-HF trial,25 the target dose was 200 mg b.i.d. This relatively high dose was achieved in 64% of the patients, and the mean dose in the trial was 159 mg. There is scant data for lower doses in terms of mortality end points. In fact, there is very little data with lower doses for any end point. Yet, in clinical practice, many physicians do not titrate β blockers to the doses achieved in the clinical trials. Are these patients getting any benefit with lower doses of β blockers? In one study19 a relatively small dose of metoprolol tartrate (25 b.i.d.) was used and achieved comparable results to those patients receiving carvedilol 25 mg b.i.d. The Multicenter Oral Carvedilol Heart Failure Assessment (MOCHA) trial was a prospectively designed β blocker dose-response study29 that did not achieve its prespecified primary end point of exercise capacity. Thus, any conclusions from MOCHA must be taken with appropriate caution. In MOCHA, heart-failure patients were randomized to receive carvedilol 6.25, 12.5, or 25 mg b.i.d. or placebo. The secondary end point of ejection fraction showed that higher doses of carvedilol had a greater effect on ejection fraction than the lower doses, but even 6.25 mg b.i.d. had a beneficial affect on ejection fraction compared with placebo. There were similar beneficial effects seen with hospitalization and mortality. A retrospective analysis of the MERIT-HF database30 compared the efficacy of metoprolol succinate in patients who were taking 100 mg or less of extended-release metoprolol vs. those taking more than 100 mg. Interestingly, both groups showed a similar reduction in heart rate to 67 bpm after up-titration. Investigators probably limited their up-titration based on the clinical and heart rate responses. It appears that individual patients respond differently to β-blockade dose and perhaps consideration should be given to a target or change in heart rate as the goal of up-titration. The analysis by Wikstrand et al.30 showed comparable risk reduction in mortality in the high- and low-dose group. Since the MOCHA study29 did not achieve its primary exercise endpoint and the MERIT-HF analysis30 was retrospective, the definitive dose response answer awaits further study. In fact, a prospective double-blind positive controlled dose response study using an eightfold dose range of extended release metoprolol succinate (25, 50, 100, and 200 mg q.d.) for NYHA class II–III heart failure patients is now enrolling patients (A phase IV, multicenter, double-blind, randomized, parallel group, dose-response study to assess the effect of extended release metoprolol succinate on cardiac remodeling in symptomatic chronic heart failure subjects with left ventricular dysfunction [MD-CHF]). For the time being, one should aim for target doses achieved in clinical trials, but if during up-titration patients develop side effects such as fatigue, worsening heart failure, symptomatic hypotension, or symptomatic bradycardia, the dose should be reduced to the previously tolerated dose. One can reasonably conclude with the present data that some β blocker is better than none. Perhaps the greatest challenge and also the greatest impact would be to increase the use of β blockade in the appropriate patient populations. The current estimate is that less than 20% of appropriate heart failure patients are taking β blockers at any dose.31 Physician, nurse, and patient education must be improved to increase the usage of this life-saving therapy.32 Clearly β blockers are indicated in nearly all patients with NYHA functional class II–III heart failure. The ongoing Reversal of Ventricular Remodelling with Toprol-XL (REVERT) trial using extended-release metoprolol succinate may help clarify the utility of β blockers in patients with asymptomatic LV dysfunction (NYHA class I). The Carvedilol Post-Infarct Survival Control in LV Dysfunction (CAPRICORN) study33 has demonstrated the mortality benefit of carvedilol in patients after a recent myocardial infarction with asymptomatic left ventricular dysfunction in addition to treatment with angiotensin-converting enzyme inhibitors.34 Perhaps the biggest hurdle to overcome in increasing usage of β blockers is the reasonable concern about the acute negative inotropic35 and hemodynamic36 effects of β blockers during initiation. However, considering the significant mortality and morbidity data, an approach of starting with a very low dose and a gradual up-titration over several weeks must be encouraged. Although the two US Food and Drug Administration-approved β blockers for heart failure (carvedilol and metoprolol succinate) have different pharmacological properties beyond β-1 blockade, there is no data supporting the preferential use of one agent over the other.37 The currently available data suggests that the primary benefit is from β-1 blockade and supports the usage of either carvedilol or metoprolol. Perhaps the different mechanisms and nuances between metoprolol and carvedilol as detailed in this issue of CHF are more of academic interest than of clinical significance. Is it akin to comparing a hockey team with muscle and an intimidating style to that of a team with skating finesse? Is it similar to the debate in baseball comparing a team with a formidable pitching staff and defense to that of a home run-producing power offense? Both styles of hockey and baseball teams win games. What is crucial is to increase the overall usage of β blockers in appropriate heart-failure patients. To achieve this, it is important that health care providers are educated regarding the appropriate dosing and titration of β blockers; efforts should be made to further encourage the use of these life-saving medications in clinically stable patients with mild-to-moderate heart failure.