Slowing the progression of heart failure.

Publication of the results of the U.S. carvedilol programme' has created an international dilemma affecting patients, physicians and drug regulatory agencies. The data presented in the New England Journal of Medicine of 23 May 1996 were from the overall study population of 1094 patients with symptoms of chronic heart failure, a left ventricular ejection fraction <O35, and the ability to perform a corridor walk test. All were on what was viewed as optimal therapy with diuretics and a converting enzyme inhibitor. They were randomly assigned to treatment with carvedilol or placebo. During an average of 6 months of follow-up 7-8% of the placebo group but only 32% of the carvedilol group died, a 65% mortality reduction with 95% confidence intervals from 39 to 80%. This statistically remarkable benefit was also reflected in a 27% reduction in the need for hospitalization for cardiovascular causes. Carvedilol is a non-selective /?-adrenoceptor blocker that exerts a vasodilator effect through a, receptor blockade and has also been shown to have potent antioxidant properties. Should the druginduced reduction in short-term mortality in this population of patients with mild to moderate symptomatic heart failure be taken as evidence that /?-blockers prolong life? Do these data suggest that all patients with heart failure should be treated with carvedilol if tolerated? Is there adequate evidence of efficacy and safety to justify approval of this drug for treatment of heart failure? The complexity of these questions relates not only to one's interpretation of the data from the carvedilol studies but also to our emerging understanding of heart failure. The syndrome appears to involve two largely independent processes: (1) dysfunction and remodelling (dilatation) of the left ventricle, and (2) a symptom complex characterized by congestion and exertional fatigue. Treatments for heart failure have traditionally been evaluated on the basis of their efficacy in short-term symptom relief, for which exercise testing and, more recently, quality of life questionnaires have been used as guides. However, an equally important and perhaps more modifiable goal would be slowing of progression of the syndrome. This is probably best assessed by monitoring the structural and functional changes in the left ventricle. These anatomical and physiological changes may well be accompanied by long-term effects on symptoms and the need for hospitalization, but early drug-induced changes in exercise tolerance and quality of life, indicative of a therapeutic effect on symptoms, is quite distinct from a therapeutic effect to delay or prevent the worsening of the syndrome over time. The dissociation between symptoms and left ventricular structure was probably best exemplified in the SOLVD Prevention Trial which sought asymptomatic individuals with dilated, dysfunctional left ventricles (ejection fraction <35%) for randomization to enalapril or placebo therapy'. Over 4000 patients were recruited into this trial, and their ejection fractions (mean 0-28) largely overlapped with those in the SOLVD Treatment Trial with symptomatic heart failure (mean 0-25). The success of enalapril in reducing the rate of development of symptomatic heart failure in the Prevention Trial was taken as evidence that converting enzyme (ACE) inhibitor therapy should be used to prevent congestive heart failure. More significantly, however, ACE inhibitors in the SOLVD and post-myocardial infarction SAVE Trials' delayed the progressive structural and/or functional abnormality in the left ventricle by resulting in a higher ejection fraction in the treated than in the placebo groups. Even in the absence of an early symptomatic benefit, this physiological effect might be expected to result in a long-term reduction in morbid events and mortality. In dealing with a therapy for heart failure, therefore, we should define whether our goal is shortterm relief of symptoms or long-term slowing of the rate of progression of the disease. Therapeutic use of /?-blockers for heart failure has had a long and controversial history. Early trials with metoprolol in Sweden were uncontrolled observational studies that suggested a long-term benefit on morbidity and mortality'. Even in these early trials it was emphasized that symptoms often worsened in the early weeks or months after initiating the gradually titrated regimen. The failure of subsequent short-term placebo-controlled studies to demonstrate symptom relief was the expected outcome with a drug aimed at slowing progression rather than at short-term improvement in exercise tolerance. However, the rather unimpressive and unpersuasive