Skip to content

Perspectives in cardiac failure therapeutics

Perspectives – medication for heart failure.

The obvious drawback to this image/comparison is the weakness of the spironolactone data in terms of absolute numbers of events compared to the relatively large trials on the other agents…

Spironolactone

RALES, the pivotal trial1 that released spironolactone back into the therapeutic armamentarium for cardiac failure, warrants some comment. It involved a highly selected group of patients (73% males of average age 65 years, with a mean ejection fraction of 26%, a serum creatinine less than 221 micromol/l and potassium less than 5 mmol/l), and about half of whom had an ischaemic aetiology.

The mortality rate in the control group was 46% and in the spironolactone group 35%, yielding an 11% ARR (95% CI 6-16%), hence NNT 9 (6 to 16) at 2 years. The prevalence of serious hyperkalaemia was 1% in the control group and 2% in those given spironolactone.

It is still not entirely clear how generalisable the results will be. In particular, the not uncommon extension of the prescription of spironolactone to patients with cor pulmonale or CCF with renal impairment cannot be supported by this trial.

A time series analysis of individuals over the age of 65 in Ontario2 showed a rise in both the rate of hyperkalaemia-associated hospitalisation (2.4 per 1000 patients in 1994 to 11.0 per 1000 patients in 2001) and hyperkalaemia associated death (0.3/1000 in 1994 to 2.0/1000 in 2001) after the publication of the RALES study. Particular caution is warranted if patients are co-prescribed beta-blockers, and in older patients with worsening heart or renal failure or diabetes3

ACE inhibitors.

The SOLVD trial4 looked at the efficacy of enalapril in patients with Grade II and III heart failure and showed a significant survival advantage. Median follow-up time was 41 months, but for comparative purposes, the mortality reduction benefit can be expressed over one year for ACE inhibitors, beta-blockers and spironolactone5 as 1.84% (NNT 55), 3.74% (NNT 27) and 6.9% (NNT 15) respectively. This rather neat halving of NNTs from one agent to the next is confounded by variable baseline mortality – the control one year mortality was 11% for the ACE inhibitor and beta-blocker trials, and 23% for spironolactone.

(Benefit here is expressed over one year and after a bit of statistical manipulation; the original RALES data giving a NNT of 9 for spironolactone was for two years of treatment.)

Beta-blockers

Beta-blockers offer a survival advantage to patients with heart failure. However they should only be given to patients who have been adequately treated with ACE inhibitors and diuretics plus spironolactone. In most of the trials between one and two thirds of patients were also on digoxin. Many patients seen in a resource-constrained practice are dipping in and out of severe fluid overload due to difficulty in accessing medication and in adhering to sometimes overly complicated regimens. In such individuals, the attempted addition of a beta-blocker is unwise.

Pragmatic trials looking at what happens when one tries to apply the results of the carvedilol and metoprolol survival data outside the RCT environment are starting to appear. One6 looked at a cohort of 100 patients in a London district general hospital. Of these, 22 were already on a beta-blocker. Of the remainder, 22% had contra-indications or refused therapy with a beta-blocker, 9% failed the test dose of 3.125 mg of carvedilol 12 hourly, 23% could only tolerate this dose, 25% reached 6.25 mg 12 hourly, 15% reached 12.5 mg 12 hourly and only 5% reached the target dose.

The other way of looking at this is that about two thirds of patients eventually got at least some beta-blockade, and even low doses may help. The authors estimated that this cohort of 100 patients took up 155 hours of work time, or a mean of 2.5 hours per patient initiated.

Digoxin

When digoxin was given in conventional doses, it had no effect on mortality in the now elderly (1997) DIG trial, although it may have reduced hospitalisation7. That trial has been criticised in the light of subsequent concerns that the proportion of patients with serum digoxin concentrations above 1.0 ng/ml may have masked benefit in those treated with more recently recommended lower doses. This concern was substantiated by a post hoc review of the trial data showing some benefit in a highly selected group of patients8. All analyses of this type should be treated with extreme circumspection, but the results, for what they are worth, showed an absolute risk reduction (mortality) of 6.3% for those with digoxin levels between 0.5 and 0.8 ng/ml, and an increase in mortality of 11.8% in those with digoxin levels more than 1.2 ng/ml.

It is tempting to over-emphasize this highly significant advantage over placebo, but it is important to remember the selection criteria – digoxin levels of 0.5-0.8 ‘selected’ for a group of men (women excluded) who were younger, and fatter, and had better renal function and a considerably lower nitrate use. These differences (and others) were adjusted for in a proportional hazards model and mortality differences still seemed real. This does not imply causality – higher digoxin levels may just have been markers for higher risk attributable to another unmeasured risk factor.

Another post-hoc review9 looked at the percentage patients developing worsening cardiac failure on treatment, and found that this occurred in 39% of patients on a diuretic alone, 25% on an ACE inhibitor and a diuretic, 19% on digoxin and a diuretic, but only 5% (p < 0.001 relative to the other groups) in those given all three agents.

More recently, the AFFIRM investigators reported yet another post-hoc digoxin analysis10 which has cast considerable doubt on the ongoing use of this agent. This study was designed to compare rate and rhythm control, and digoxin use was not randomised. There was increased mortality associated with digoxin use (HR 1.41, 95% CI 1.19 to 1.67) in this study of 4060 patients with follow-up out to 6 years. However, more than half the patients in this study did not have heart failure, so direct extrapolation is not easy. In those with reduced ejection fraction (<40%) the mortality disadvantage of digoxin was still seen.

The take home message with our current state of knowledge may be that adding digoxin is not essential in the management of cardiac failure and may cause harm, but it may still have a role in selected patients.

There is now some evidence that comprehensive discharge planning can reduce re-admission rate in older patients with CCF11 – mortality was reduced by 3% (ARR), and re-admission rate by 8%. Most of this improvement was attributed to provision of day hospital services (mortality ARR 16%) Increased hospital clinic follow-up was associated with worsening mortality (1% absolute increase.)

While some of these differences look quite encouraging, none were statistically significant other than the 8% reduction in total re-admissions, and the mortality advantage of day hospital services (but this last was derived from a single trial.) Overall, and rather more soberingly, a large US study12 looked at the impact of a number of strategies in reducing readmission in CCF patients, and found only modest impact of most:

Intervention efficacy CCF

  1. Pitt B, Zannad F, Remme WJ, et al. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. N Engl J Med. 1999;341:709-17 ↩

  2. Juurlink DN, Mamdani MM, Lee DS, et al. Rates of hyperkalaemia after publication of the randomised aldactone evaluation study. N Engl J Med. 2004;351:543-51 ↩

  3. Schepkens H, Vanholder R, Billiouw J, et al. Life-threatening hyperkalaemia during combined therapy with angiotensin-converting enzyme inhibitors and spironolactone: an analysis of 25 cases. Am J Med. 2001;110:438-41 ↩

  4. SOLVD investigators. Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. N Engl J Med. 1991;325:293-302 ↩

  5. Attia J, Page J, Heller R, et al. Impact numbers in health policy decisions. J Epidemiol Community Health. 2002;56:600-605 ↩

  6. Mehta PA, McDonagh S, Poole-Wilson PA, et al. Heart failure in a district general hospital: are target doses of beta-blockers realistic? Q J Med. 2004;97:133-9 ↩

  7. The Digitalis Investigation Group. The effect of digoxin on mortality and morbidity in patients with heart failure. N Engl J Med. 1997;336:525-33 ↩

  8. Rathore SS, Curtis JP, Wang Y, et al. Association of serum digoxin concentration and outcomes in patients with heart failure. JAMA 2003;289:871-78 ↩

  9. Young JB, Gheorghiade M, Uretsky BF, et al. Superiority of “triple” drug therapy in heart failure: insights from the PROVED and RADIANCE trials: Prospective Randomized Study of Ventricular Function and Efficacy of Digoxin, Randomized Assessment of Digoxin and Inhibitors of Angiotensin-Converting Enzyme. J Am Coll Cardiol. 1998;32:686–92 ↩

  10. Whitbeck MG, Charnigo RJ, Khairy P, et al. Increased mortality among patients taking digoxin – analysis from the AFFIRM study. European Heart J. 2012: doi:10.1093/eurheartj/ehs348 ↩

  11. Phillips CO, Wright SM, Kern DE, et al. Comprehensive discharge planning with post-discharge support for older patients with congestive heart failure. JAMA 2004;291:1358-67 ↩

  12. Bradley EH, Curry L, Horwitz LI, et al. Hospital strategies associated with 30-day readmission rates for patients with heart failure. Circ Cardiovasc Qual Outcomes. 2013;6:444-450 ↩