New Heart Failure Clinical Trials
Synopsis: Several studies were reported at the Second Annual Scientific Meeting of the Heart Failure Society of America that should influence physician practice in the care of patients with left ventricular dysfunction and heart failure.
Source: The Second Annual Scientific Meeting of the Heart Failure Society of America, Boca Raton, FL. September 13-16, 1998.
This important trial evaluated the effects of bisoprolol, a selective beta 1 beta-blocker, in patients with heart failure. The trial was stopped prematurely at the second interim analysis because of positive results in the beta blocker arm. Approximately 2600 patients from throughout Europe with Class III or IV congestive heart failure were slowly up-titrated with bisoprolol or placebo over a period of several months. All were on an angiotensin converting enzyme (ACE) inhibitor and diuretics. Entry criteria included an ejection fraction (EF) of less than 35%; 16% of the patients were Class IV. Eighty percent were male, and more than 50% had coronary artery disease. The primary end point was all-cause mortality; a variety of traditional secondary end points were assessed. At the time the trial was stopped, all-cause mortality had decreased in the beta blocker group by 32% (P = 0.0005); death rates were 17.3% placebo vs. 11.8% bisoprolol, with a rate of 12% per year in the placebo arm and 8.2% in the beta blocker cohort. Average follow-up at trial cessation was 1.4 years. There was a 45% decrease in sudden death and a slight favorable trend in deaths from heart failure or unknown causes. There were no significant differences in outcome in subjects with an ischemic etiology (50% reduction in deaths) or different functional class. Total and heart failure hospitalizations were decreased in the beta blocker group. Virtually all secondary end points were positively affected, including in-hospital deaths. Withdrawal rates were 15% for both placebo and bisoprolol. In summary, CIBIS-2 resulted in a 32% reduction in all-cause mortality, 45% reduction in sudden death, 30% reduction in hospitalization for CHF, and 15% reduction in all-cause hospitalization. No significant adverse reactions occurred. The authors conclude that only 25 patients would need to be treated with bisoprolol to save one life.
COMMENT BY JONATHAN ABRAMS, MD
This important study confirms recent meta-analysis (Circulation, 1998;98:1184) demonstrating an advantage in death or heart failure hospitalization as well as EF in more than 3000 patients receiving a beta blocker who have congestive heart failure. While the mortality rates in CIBIS II suggest that these patients may have been less sick than traditional Class III-IV classification, the data are concordant with the recent carvedilol studies as well as outcomes in a number of small beta blocker trials. Thus, it would appear that all patients who have congestive heart failure with substantial depression of EF should be given a beta blocker unless there are contra-indications. Certainly, this is an attractive policy for stable Class II-III subjects. The question as to whether selective, non-selective, or vasodilator-beta blockers are superior is unresolved and awaits the results of ongoing trials (BEST, COMET, COPERNICUS). Very recently, the MERIT-HT study was stopped because of a major benefit of long-acting metoprolol in 4000 cases of II-IV subjects with an EF less than 40%. The data are not available yet, but this study, along with CIBIS II and the carvedilol trials, underscores that beta blockers clearly increase survival in heart failure with impaired LV systolic function.
This trial was an international evaluation of the T-channel calcium blocker mibefradil, recently removed from the market in the United States because of drug interactions and other toxicity. This promising calcium blocker had received considerable attention, and the results of the MACH-I trial were eagerly anticipated as to whether a calcium blocker could benefit some patients with depressed LV function and heart failure. The primary study end point was all-cause mortality, with a variety of secondary end points. Subjects with an EF of less than 35%, on diuretic and ACE inhibitor, functional Class II-IV, were enrolled after one month of up-titration. Approximately 2600 individuals were enrolled with a mean EF of 25%. Two-thirds had coronary disease; the majority were Class III (10% Class IV, 25% Class II). The study showed a nonsignificant increase in deaths of approximately 11% in the mibefradil group, with an average follow-up of 580 days. There was an early increase in mortality with the calcium channel blocker. A number of drug interactions were evaluated because of the effects of mibefradil on the cytochrome P-450 system; individuals (mostly women) who received the calcium blocker and took amiodarone or other anti-arrhythmics showed a substantially higher death rate.
COMMENT BY JONATHAN ABRAMS, MD
This trial, while already of historic interest, does add to the database regarding whether there is a "safe" calcium channel blocker for heart failure. PRAISE II is asking whether amlodipine may be better than placebo in patients with dilated cardiomyopathy. Data from PRAISE I and several felodipine trials suggest that both amlodipine and probably felodipine are at least safe in these patients. Mibefradil had demonstrated an excellent profile in animal models and showed promise in humans with left ventricular dysfunction in acute studies. Once again, calcium blockers have been a major disappointment for long-term therapy of heart failure. The interaction with many compounds metabolized through cytochrome P-450 was unsuspected and resulted in the withdrawal of mibefradil in the United States after 10 months on the market for angina pectoris. The role of calcium blockers as therapeutic agents for congestive heart failure remains an unfulfilled promise; MACH I is another sad chapter for new and "promising" agents for heart failure.
Recognition that ACE inhibition in patients with congestive heart failure does not provide sustained suppression of angiotensin II is well recognized. What is less known is that aldosterone levels rise over time, potentially resulting in important losses in potassium, particularly in patients on a low-sodium diet. Preliminary data indicate that both ACE inhibitors and angiotensin receptor blockers do not indefinitely suppress aldosterone levels. Other data link high aldosterone to mortality (Consensus I). Aldosterone is known to not only promote potassium loss, but it is also associated with ventricular fibrosis, sodium retention, and a variety of other actions that could be adverse in patients with depressed LV function. The RALES Trial was designed to see whether spironolactone would be beneficial in heart failure by preventing elevations in aldosterone. Sixteen hundred patients with Class III-IV heart failure were enrolled for a projected three-year follow-up. All patients were on an ACE inhibitor and a diuretic, and many were on digitalis. Spironolactone was up-titrated, beginning with 25 mg; serum potassium was carefully monitored. The target dose was 75 mg daily. Primary end point was all-cause mortality. The study was stopped by the Data Safety and Monitoring Board on August 24th, 1998, because of a significant benefit in the sprironolactone arm. No data are available at this time.
COMMENT BY JONATHAN ABRAMS, MD
Much credit must be given to Bert Pitt, MD, for the conceptualization of this trial. Details will be released at the upcoming American Heart Association meeting. In a sick population of heart failure patients, there clearly was sufficient benefit provided by spironolactone that resulted in early termination of the trial, approximately 16 months before projected completion. Hyperkalemia was noted in approximately 22-28% of individuals who reached the 50 mg or 75 mg dose, respectively. Nevertheless, the study protocol allowed for careful dosage alterations linked to potassium monitoring. The actual results will be of great interest. It may be that spironolactone, an old and little used agent, may provide an important new therapeutic adjunct for heart failure.
This study, initially reported at the American College of Cardiology meeting in March 1998 (Clin Cardiol Alert 1998;17:34), was presented in more detail by Milton Packer, MD. To reiterate, ATLAS asked the question whether high-dose ACE inhibitor was substantially better than low dose cohort. Class III patients with ejection fraction of less than 30% were followed for approximately four years. The mean high dose of lisinopril averaged 33.5 mg vs. 3.5 mg in the low dose. Primary end point was all-cause mortality. There was an 8% trend toward increased survival in the high-dose group, which was not significant. Cardiovascular deaths almost achieved significance (P = 0.07), whereas all-cause death or hospitalization was significantly reduced in the high-dose group (84 vs 80%). Other combined end point results were positive in the high-dose group, as were hospitalizations. Side effects were similar. Recurrent heart failure hospitalization was reduced by 24%. Packer concluded that high doses of ACE inhibitors would substantially reduce hospital admissions and death from congestive heart failure substantially.
COMMENT BY JONATHAN ABRAMS, MD
As previously discussed in Clinical Cardiology Alert, the major problem with this study was the use of an extremely low dose of lisinopril in one arm and, thus, setting up an artificial construct. Most patients today are treated with 10-20 mg of lisinopril. Clearly, low doses are not as effective as higher doses. Nevertheless, the study did not demonstrate a robust decrease in any end point, which is somewhat surprising considering the marked difference in dose between the low- and high-dose groups. In general, physicians should continue to use higher doses of ACE inhibitors, as many surveys document underdosing.