Clinical trials update from the Heart Failure Society of America and the American Heart Association meetings in 2008: SADHART-CHF, COMPARE, MOMENTUM, thyroid hormone analogue study, HF-ACTION, I-PRESERVE, β-interferon study, BACH, and ATHENA

Abstract

This article provides information and a commentary on trials relevant to the pathophysiology, prevention, and treatment of heart failure presented at the Heart Failure Society of America and the American Heart Association meetings in 2008. Unpublished reports should be considered as preliminary, as analyses may change in the final publication. (i) SADHART-CHF showed no difference in outcome for heart failure patients with depression treated with sertraline compared with placebo. (ii) A controlled release carvedilol formulation showed similar LV haemodynamic effects to the standard carvedilol formulation in the COMPARE study. (iii) A post hoc analysis of the MOMENTUM study suggested that patients with less severe heart failure may be more likely to benefit from a continuous aortic flow augmentation device. (iv) A thyroid hormone analogue was poorly tolerated in patients with heart failure. (v) HF-ACTION showed that exercise training is safe and offers modest clinical benefits in patients with heart failure. (vi) Irbesartan failed to improve outcomes in patients with preserved ejection fraction in the I-PRESERVE study. (vii) A phase II study of beta-interferon administration in patients with dilated cardiomyopathy showed encouraging results. (viii) The BACH study showed that mid-regional pro-adrenomedullin was more accurate than BNP or NT-proBNP at predicting outcome at 90 days in patients with acute heart failure. (ix) A secondary analysis from ATHENA showed a reduction in cardiovascular hospitalizations and strokes for patients with atrial fibrillation receiving dronedarone compared with placebo.

SADHART-CHF: safety and efficacy of sertraline for depression in patients with chronic heart failure trial

Presented by Christopher O’Connor from Durham, NC, USA.

Major depression is common in patients with heart failure and is associated with a poor prognosis.^1–3 Sertraline is a selective serotonin reuptake inhibitor (SSRI), which appears safe and effective for the treatment of depression after an acute coronary event.⁴ SADHART-CHF is the first large, randomized, controlled trial to examine the effects of an SSRI on depression and cardiovascular outcome in patients with heart failure and concomitant major depression.⁵

A total of 469 patients aged ≥45 years, with NYHA class II–IV heart failure, left ventricular ejection fraction (LVEF) ≤ 45%, Beck Depression Inventory (BDI) score ≥10, and meeting the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) criteria for major depressive disorders were randomized to treatment with either sertraline (n = 234) or placebo (n = 235). Many of the patients had recently been discharged after a hospitalization for worsening heart failure. Treatment was initiated at 50 mg per day and up-titrated to a maximum of 200 mg. The first 12 weeks of treatment were double-blind, followed by an open-label treatment (with either sertraline or citalopram) for up to 2.5 years. Patients in both treatment groups also received nurse-facilitated support.

The primary endpoints were change in depression score at 12 weeks and cardiovascular status, scored as improved, unchanged, or worsened, at 12 weeks and during long-term follow-up.

The mean age of patients was 63 years, and mean LVEF was 30%. Approximately 80% of patients were receiving combination treatment with both an ACE-inhibitor and a beta-blocker. The mean BDI score at baseline was 19.1 and the Kansas City Cardiomyopathy Questionnaire (KCCQ) score was 45.9.

Over the 12 week treatment period, depression in both groups improved markedly, but there was no difference in the change in depression scores or clinical outcome between the treatment groups. The Hamilton Depression Rating Scale and the KCCQ scores improved similarly in both groups. There was no difference in the incidence of death, cardiovascular death, or hospitalization for worsening heart failure (Table 1). The dropout rate was high in both treatment groups.

Table 1.

Outcomes at 12 weeks in the SADHART study

	Sertraline (n = 234)	Placebo (n = 235)	P-value
Death (n)	18	15	ns
CV death (n)	15	10	ns
Hospitalization for worsening HF (n)	19	30	0.1

ns, not significant.

There are several potential interpretations of the results. It is possible that sertraline is ineffective. Alternatively, the more intensive nursing care could have been enough to improve many patients, suggesting that sertraline should be re-studied in patients who have failed to respond to general measures. Many of the patients were enrolled during or shortly after hospitalization for worsening heart failure, and spontaneous improvement might be anticipated among the survivors of such an event. Finally, the study may have been under-powered. It would require a study about 6 times larger and 12 times longer to confirm whether or not the trend to a lower rate of hospitalization for heart failure was real.

The results of another placebo controlled study (MOOD-HF), which aims to evaluate the effects of the SSRI escitalopram on outcomes (including morbidity and mortality) in depressed patients with NYHA class II–IV heart failure and impaired left ventricular systolic dysfunction over a 1–2-year follow-up period, may provide further answers.⁶

COMPARE: comparison of the effects of carvedilol controlled release and carvedilol immediate release on left ventricular ejection fraction in patients with heart failure

Presented by Barry Greenberg from San Diego, CA, USA.

For patients with heart failure, carvedilol is currently given as an immediate release (IR) formulation, administered twice daily. A controlled release (CR) preparation could increase convenience and improve compliance by reducing the frequency of dosing to once per day. The COMPARE study was undertaken to compare a once-daily slow-release formulation of carvedilol, initiated at 10 mg per day and titrated to a maximum of 80 mg per day vs. conventional IR carvedilol initiated at 3.125 mg b.i.d. and titrated to a maximum of 25 mg b.i.d.⁷ Efficacy was assessed using changes in echocardiographic variables, haemodynamics, and biomarker levels. The primary endpoint was change in LV end-systolic volume index (LVESVI). COMPARE was a non-inferiority study requiring a sample size of 250 complete sets of patient data.

Patients with stable chronic heart failure and an LVEF < 40% were randomized to treatment with either carvedilol CR or carvedilol IR for up to 24 weeks. Suitable patients were required to be beta-blocker naïve (defined as none within 42 days) but on an ACE-inhibitor or angiotensin II receptor blocker (ARB) unless intolerant. Patients were stratified according to the presence of ischaemic heart disease (IHD).

Three hundred and eighteen patients were randomized, 153 patients to carvedilol CR, and 165 to carvedilol IR. Of these, 70% were men, the majority were white, 32% had IHD, 21% had diabetes, and 64% were on a loop diuretic. About 80% of patients in both groups achieved the target dose.

The effects of the two different carvedilol formulations on cardiac function are summarized in Table 2. The large increases in LVEF probably reflect the fact that few patients with IHD were enrolled. Patients with IHD have more LV scar that cannot recover contractility with beta-blockers and so have a smaller improvement in LV function.⁸ Only 4 deaths and 12 heart failure-related hospitalizations occurred in the study; so, not surprisingly, no differences in clinical outcomes could be observed. The reason for the differences in blood pressure remains unexplained but might relate to differences in the time-course of alpha-adrenergic receptor blockade between agents.

Table 2.

Effect of carvedilol preparations on cardiac function in the COMPARE study

	Carvedilol CR		Carvedilol IR		P-value
	Baseline	24 weeks	Baseline	24 weeks
LVESVI (mL/m2)	69	48	63	45	ns
LVEDVI (mL/m2)	96	76	90	72	ns
LVEF (%)	30	38	31	39	ns
SBP (mmHg)	117	124	116	117	<0.0005
BNP (% reduction from baseline)	36		36		ns

ns, not significant.

MOMENTUM: multicentre trial of the Orquis^® medical cancion system for the enhanced treatment of heart failure unresponsive to medical therapy

Presented by Mandeep Mehra from Baltimore, MD, USA.

MOMENTUM has been reported previously in this journal⁹ and the primary manuscript published recently.¹⁰ Briefly, this was a study of a continuous aortic flow augmentation CAFA device for the management of severe decompensated heart failure with incipient cardiogenic shock. The post hoc analysis showed that patients who were less sick, as evidenced by NT-proBNP below about 5000 pg/mL, were more likely to benefit. The concept that greatest treatment benefit for patients with heart failure may be observed in patients at intermediate rather than high risk seems to be a consistent message from recent studies including SCD-HeFT, CARE-HF, and CORONA.¹¹

Thyroid hormone analogue for heart failure

Presented by Steven Goldman from Tucson, AZ, USA.

Thyroid hormone supplementation has been suggested as a treatment for heart failure;^12–15 benefits may be mediated by increased cardiac output, vasodilatation, and neo-angiogenesis.

Of the 86 patients enrolled in the study, 57 were randomized to 3,5-diiodothyropropionic acid (DITPA) and followed up using non-invasive haemodynamics (bio-impedance) for 6 months. DITPA was not well tolerated at the dose given (44% withdrew from treatment) and the study was stopped by the Veterans Association review board due to adverse effects and futility for clinical outcome. Patients felt worse on therapy despite substantial and persistent improvement in cardiac output. Patients lost about 5 kg in weight compared with placebo and had a marked decline in cholesterol, suggesting the development of drug-induced cachexia, presumably induced by an increase in metabolic rate. Identifying a dose that does not cause progressive weight loss that might still stimulate new vessel growth might be beneficial, although there could be a risk of increased tumour growth. There could be a role for this agent in the treatment of morbid obesity should it be shown to be safe.

HF-ACTION: efficacy and safety of exercise training as a treatment modality in patients with chronic heart failure: results of a randomized controlled trial investigating outcomes of exercise training

Presented by David Whellan from Philadelphia, PA, USA and Kathryn Flynn from Durham, NC, USA.

Exercise training reduces symptoms in patients with chronic heart failure but evidence for an effect on morbidity and mortality is limited to a number of small studies and meta-analyses.^16,17 Although it is widely accepted that exercise training is beneficial in patients with heart failure,¹⁸ it is underused in clinical practice. HF-ACTION is the first large, randomized, controlled trial to evaluate the effects of exercise training in patients with heart failure.¹⁹

Patients with chronic NYHA class II–IV heart failure, with an ejection fraction of ≤35%, and on optimal therapy who were considered to be able to exercise safely were randomized to either exercise training (n = 1159) or usual care (n = 1172). Patients randomized to usual care had an evaluation of cardiopulmonary exercise capacity and were advised to do more exercise but were not offered exercise training within the protocol. The exercise intervention aimed at achieving 90 min of exercise per week for the first 12 weeks, increasing to 120 min per week thereafter. The primary endpoint was all cause death or hospitalization. Owing to the nature of the intervention, blinding of treatment allocation was not possible.

The mean age of patients was 59 years, 29% were females and the mean ejection fraction at baseline was 25%. The majority of patients were receiving treatment with ACE-inhibitors (92%) and beta-blockers (95%).

Adherence to the exercise training protocol was reported to be suboptimal, with most patients failing to achieve the target level of exercise. Despite this, there was an improvement in exercise duration and peak VO₂ in the exercise training group at 12 months compared with usual care (P < 0.0001 for both), but there was no difference in 6MWT distance.

There was no difference in the unadjusted incidence of all-cause mortality or all-cause hospitalization at 2.5 years of follow-up (HR 0.93, P = 0.13); however, when adjusted for key prognostic variables the difference was significant (HR 0.89, P = 0.03).

The results of an HF-ACTION sub-study, which were also reported at the AHA meeting, showed that exercise training was associated with an early and sustained improvement in quality of life, assessed using the KCCQ. At 3 months, patients in the exercise group showed an average five-point gain on the 100-point KCCQ scale, compared with only a three-point gain for the usual care group (P = 0.005). The proportion of patients achieving at least a five-point gain on the KCCQ scale was higher in the exercise group both at 3 months (54% vs. 28%) and 1 year (53% vs. 33%) (P ≤ 0.0001 for both).

The results of HF-ACTION demonstrate that exercise training is safe and offers modest clinical benefits in patients with heart failure. The study also highlights the difficulties associated with ‘dosing’ of this kind of treatment regime. Not only is it difficult to ensure that patients on active treatment comply with the exercise programme, it is also difficult to restrict exercise in the control group, and it is likely that many of these patients took up the suggestion to do more exercise. Ultimately, people will generally do only the amount of exercise they feel comfortable with. Those who feel able to exercise will do so and those who do not will not. The greatest paradox is that exercise seems to improve well-being but getting patients to sustain the effort is difficult,²⁰ which suggests that exercise does not have that great an effect on quality of life in many of these patients. Many frail, older patients are unable to exercise substantially. Whether low levels of exercise can help retain mobility and improved quality of life is uncertain.

I-PRESERVE: irbesartan in heart failure with preserved ejection fraction

Presented by Peter Carson from Washington, DC, USA.

Some studies suggest that up to 50% of patients with heart failure have a preserved ejection fraction; however, few treatments have been specifically evaluated for efficacy in this patient cohort. I-PRESERVE was designed to study the effect of the angiotensin II receptor blocker irbesartan in patients with an ejection fraction of 45% or more.^21,22

A total of 4128 patients aged 60 years or older with NYHA class II–IV heart failure and an ejection fraction ≥ 45% were randomized to treatment with either irbesartan 300 mg per day or placebo. Over a mean follow-up period of just over 4 years, there was no difference in the incidence of the primary composite endpoint (all-cause mortality or CV hospitalization) between the treatment groups. Secondary endpoints and subgroup analyses also suggested no benefit of irbesartan over placebo in this patient group. The findings highlight the urgent need for a better understanding of the mechanisms underlying the syndrome of ‘heart failure’ with preserved ejection fraction, in order to identify appropriate treatments.

Study of β-interferon

Presented by Heinz-Peter Schultheiss from Berlin, Germany.

This phase II trial assessed the effects of subcutaneous treatment with interferon beta-1b (IFNβ-1b) in patients with chronic viral cardiomyopathy. One hundred and forty-three patients with biopsy-proven chronic viral cardiomyopathy and an ejection fraction >25% were randomized to treatment with either IFNβ-1b (4 × 10⁶ IU), IFNβ-1b (8 × 10⁶ IU), or placebo, administered subcutaneously on alternate days for 24 weeks. Myocardial biopsies taken at baseline and then ∼12 weeks after the end of the treatment period showed that both doses of IFNβ-1b were more effective than placebo at achieving virus elimination or reduction in viral load (P < 0.05). There was an improvement in NYHA class compared with placebo at 12 weeks, but the effect was not maintained at 24 weeks. Quality of life and patient global assessment were both improved at 24 weeks compared with placebo. More patients in the IFNβ-1b groups reported adverse events but no safety issues were reported. In light of these findings, further studies are required to evaluate the clinical effectiveness of this potential new therapy. These should include patients with negative biopsies to discover how specific the intervention is.

BACH: biomarkers in the assessment of congestive heart failure trial

Presented by Stefan Anker from Berlin, Germany.

Previous results from BACH have been reported in this journal.²³ Adrenomedullin is a vasodilator hormone consisting of 52 amino acids. Adrenomedullin levels are increased in the presence of endothelial dysfunction, which is common in patients with heart failure and indicate a poor prognosis. Direct measurement of adrenomedullin is not feasible due to its poor stability; however, mid-regional pro-adrenomedullin (MRproADM) is a stable marker of activity.

The aim of the BACH study was to evaluate whether MRproADM was superior to BNP and NT-proBNP for predicting mortality at 90 days in patients presenting to hospital with acute shortness of breath and heart failure. Of 1641 patients recruited, approximately one-third (n = 568) was ultimately diagnosed with heart failure. The mean age of patients was 64 years, and 52% were men, 36% had a previous history of heart failure and 19% an acute myocardial infarction. Results showed that MRproADM was prognostically more accurate than BNP or NT-proBNP at predicting outcome at 90 days. NT-proBNP did not add additional prognostic information to MRproADM alone. The findings await confirmation especially in lower risk populations as most patients with a recent episode of worsening heart failure could be considered at high risk. MRproADM could be a useful alternative to NT-proBNP for risk stratification.

ATHENA: a placebo, controlled, double-blind parallel arm trial to assess the efficacy of dronedarone 400 mg b.i.d. for the prevention of cardiovascular hospitalization or death from any cause in patients with atrial fibrillation/atrial flutter

Presented at the AHA by Christian Torp-Pedersen from Copenhagen, Denmark, and by Richard Page from Washington, Seattle, USA and by Stuart Connolly from Hamilton, Canada, at the ESC.

The results of the ATHENA study, which showed a 24% reduction in the primary composite endpoint of CV hospitalization or all-cause mortality for patients with atrial fibrillation treated with dronedarone compared with placebo, have been reported in this journal previously.²⁴

A post hoc analysis of data from the ATHENA study presented at the European Society of Cardiology meeting in 2008 showed a significant reduction in the incidence of stroke for patients receiving dronedarone vs. placebo (Table 3). As in other trials of dronedarone, serum creatinine increased more in patients given dronedarone. Dronedarone is reported to inhibit renal tubular cation transport, leading to a rise in creatinine of about 18%, but is reported not to have an effect on glomerular filtration rate.²⁵ Amiodarone may have similar effects on renal function. In the ANDROMEDA study,²⁶ dronedarone was reported to be associated with an increase in mortality among patients hospitalized with new or worsening heart failure who had an LVEF < 35%. The reason for the increase in mortality is unclear, but it occurred in patients with the most severely depressed ventricular function and was mostly due to heart failure. This could reflect an adverse effect of dronedarone on ventricular function or failure to up-titrate treatments such as ACE inhibitors due to dronedarone-induced increases in creatinine. ATHENA shows that dronedarone can be used safely in patients with milder degrees of cardiac dysfunction. Understanding where the balance of benefit and risk sits for the individual patient could be difficult.

Table 3.

Stroke events and serum creatinine levels in ATHENA post hoc analysis

	Placebo	Dronedarone	HR (95% CI)	P-value
Stroke (% per year)	1.79	1.19	0.66 (0.46–0.96)	0.027
Stroke, ACS or all-cause mortality (% per year)	6.70	5.06	0.75 (0.62–0.90)	0.002
Serum creatinine (% increase)	1	4.7

The results of two secondary analyses from ATHENA were reported at the AHA meeting. The first analysis showed a 25% reduction in the incidence and a 35% reduction in the total duration of cardiovascular hospitalizations for patients with atrial fibrillation receiving dronedarone treatment compared with placebo (P < 0.001 for both). The second analysis showed that dronedarone reduced the incidence of first recurrence of atrial fibrillation for patients in sinus rhythm at baseline by 25%, and reduced the incidence of first electrical cardioversion by 32% (P < 0.001 for both).

Conflict of interest: J.G.F.C. was an investigator in the I-PRESERVE study.