Highlights of the 2011 Scientific Sessions of the Congress of the Heart Failure Association of the European Society of Cardiology

Abstract

The Annual Scientific Sessions of the Heart Failure Association of the European Society of Cardiology was held in Gothenburg, Sweden, 21–24 May 2011. Over 700 abstracts, along with many invited programs and several satellite programs were presented. Some of the late-breaking clinical trials are summarized here.

Keywords:

• heart failure
• heart rate
• mineralcorticosteroid-receptor antagonist
• randomized controlled trials
• telemonitoring

Emphasus-heart failure

The main Emphasus trial was prematurely ended because of highly significant beneficial effect of eplerenone in addition to recommended therapy in patients with mild heart failure with reduced ejection fraction (HFREF). The results have already been presented at the American Heart Association Congress in Chicago, November 2010 and published in the New England Journal of Medicine, 6 January 2011 [1].

During the congress, the results of a substudy of the Emphasus-HF trial were presented in which they assessed the effect of eplerenone on new-onset atrial fibrillation or flutter in patients with mild (New York Heart Association [NYHA] class II) HFREF. The effect of eplerenone in addition to recommended therapy was studied. This was a presepecified target of the study.

• The incidence of atrial fibrillation and flutter (AFF);

• The relation between baseline AFF and subsequent events;

• The effect of eplerenone in patients with and without AFF at baseline.

Patients with heart failure, NYHA class II, aged 55 years or over, and a left ventricular ejection fraction (LVEF) <30% (or, if between 30–35%, QRS width >130 ms) were eligible when treated with the recommended or maximally tolerated dose of angiotensin-converting enzyme (ACE) inhibitor (or an angiotensin receptor blocker [ARB] or both) and a β-blocker (unless contraindicated). Patients were included within 6 months of hospitalization for a cardiovascular reason (or, if no such hospitalization, B-type natriuretic peptide [BNP] > 250 pg/ml or NTproBNP > 500 pg/ml [males] or >750 pg/ml [females]). Exclusion criteria included serum potassium > 5.0 mmol/l, eGFR <30 ml/min/1.73 m², the need for a potassium-sparing diuretic, or any other significant comorbid condition.

The occurrence of AFF during follow-up was collected at the end of the study for all patients. Patients with new-onset AFF were defined as those without AFF at baseline who had an end point AFF during follow-up. Analyses of primary and secondary end points were performed with a Cox Propensity Hazard model with adjustment of the following prespecified baseline prognostic factors: age, eGFR, ejection fraction, body mass index, hemoglobin level, heart rate, systolic blood pressure, diabetes mellitus, history of hypertension, prior myocardial infarction and left bundle branch block or QRS duration >130 ms.

Of the 2737 patients with HFREF, 1364 were randomized to eplerenone and 1373 to placebo.

New-onset AFF occurred significantly less in the eplerenone group as compared with placebo (2.7 vs 4.5%, adjusted hazard ratio [HR]: 0.58, 95% CI: 0.35–0.96; p = 0.034) during a median follow-up of 21 months.

Patients with AFF at baseline had higher all-cause mortality and all-cause hospitalizations compared with those without AFF at baseline (42.2 vs 35.3%; p = 0.008). This effect was mainly driven by the effect on hospitalizations, because the effect on all-cause mortality only was 14.7 vs 13.7% (p = 0.789).

At baseline, 33.2% had AFF in the eplerenone group, and 35.7% in the placebo group. In both patients with and without AFF, eplerenone reduced (all-cause and cardiovascular) mortality and (all-cause and cardiovascular) hospitalizations significantly, with a nonsignificant trend of a greater effect in those with HFREF and AFF, as compared with HFREF without AFF at baseline.

After treatment with ACE inhibitors and ARBs, this study shows that eplerenone does reduce the risk of new-onset AFF in patients with heart failure. The effect of diuretics or β-blockers on new-onset AFF is unknown; these heart failure drugs have never been investigated for their effect on atrial fibrillation or flutter.

Debate about the results of the SHIFT study

Is ivabradine effective in patients with HFREF?

Although the results of the randomized Systolic Heart Failure Treatment with the If Inhibitor Ivabradine Trial (SHIFT) have already been published in the Lancet[2,3] and presented at the European Society of Cardiology Meeting in August 2010, there is an ongoing debate on how to interpret the results for clinical practice.

In short, the SHIFT trial was a randomized controlled trial conducted at 677 centers in 37 countries. Over 6500 patients with NYHA class II–IV heart failure and a LVEF <35%, a resting heart rate >70 bpm, and a heart-failure hospitalization within the previous year, were randomized to receive either placebo or ivabradine at a starting dose of 5 mg twice daily, with adjustments to achieve a resting heart rate of 50–60 bpm. All patients were receiving standard heart-failure medications according to guidelines, including ACE inhibitors or angiotensin receptor blockers, β-blockers, aldosterone antagonists and diuretics.

Over a mean follow-up of 23 months, patients taking ivabradine showed a significant 18% reduction in the HR for cardiovascular death or hospitalization for worsening heart failure, compared with the control group, driven mainly by hospitalization for worsening heart failure (Table 1).

The benefit of ivabradine appeared to go up with increasing heart rate. The HR for the primary end point was 0.93 (95% CI: 0.80–1.08) for patients commencing with a rate <77 bpm but was 0.75 (95% CI: 0.67–0.85) among those with rates of >77 bpm; the difference between the subgroups was found to be significant (p = 0.029) (Table 2).

The authors describe ivabradine as being “well tolerated” despite a 10.3% rate of bradycardia (4.6% symptomatic and 5.7% asymptomatic), which prompted withdrawal from the study by 1.5% of patients receiving the drug.

Below are some highlights of the lively debate between Michael Böhm (Saarland University Hospital, Homburg/Saar, Germany) and opponent, Heart Failure Society of America president Barry Massie (University of California and San Francisco VA Medical Center, USA).

Böhm stressed that the ivabradine could potentially extend heart-lowering drug therapy to heart-failure patients who – for whatever reason – are not on optimal β-blocker therapy. Böhm explained that “We have patients who do not want to take β-blockers, refuse to take them, and also those with side effects like hypotension or bradycardia that do not allow them to use them, resulting in a rather low level of β-blockade in patient with HFREF. These patients might benefit on ivabradine.” Importantly, however, SHIFT was designed to have ivabradine added on; ivabradine is not a substitute in patients who tolerate β-blockers. Moreover, 90% of the participants (mean age 61 years, 76% male) in SHIFT were receiving β-blocker therapy, which gives the impression that in patients with HFREF β-blocker ‘intolerance‘ is low. According to Böhm, SHIFT is a ‘proof-of-concept‘ study for the question of whether there is a causal relationship between heart rate and outcomes in heart failure, because ivabradine only reduces heart rate without any other known effect. In SHIFT, the average heart rate was reduced by 11 bpm compared with placebo, but interestingly, the higher the baseline heart rate, the more heart rate dropped off with ivabradine; it fell by a mean of 16 bpm for patients at 80–86 bpm at baseline, and by a mean of 22 bpm for those who had been at ≥87 bpm.

Massie stressed the point that the primary end point in SHIFT among patients on ivabradine “was driven primarily by the reduction in heart failure hospitalizations, accounting for 514 of 793 events. The absolute and relative risk reduction was only 1.4 and 10% (p = 0.092) for all-cause mortality, and 1.1 and 9% (p = 0.128) for cardiovascular mortality, respectively. In contrast, the absolute and relative risk reduction was 4.7 and 26% (p < 0.0001) for heart failure hospitalizations, respectively.

This brought Massie to his second point; β-blockers seem much better in reducing all-cause mortality in HFREF, probably due to effects beyond heart-rate lowering. All-cause mortality results have been consistently excellent in the large β-blocker trials (the Cardiac Insufficiency Bisoprolol Study II [CIBIS-2], the Metoprolol CR/SL Randomized Intervention Trial in Congestive Heart Failure [MERIT-HF], Carvedilol Prospective Randomized Cumulative Survival Trial [COPERNICUS], Study of the Effects of Nebivolol Intervention on Outcomes and Rehospitalisations in Seniors with Heart Failure [SENIORS]), but in the SHIFT trial ivabradine was not very powerful on this end point. For every 5-bpm reduction in heart rate with β-blockers, mortality fell 18% as shown in the meta-analysis of β-blocker trials of McAllister and colleagues [4]. Reducing heart rate beyond what is currently accomplished with β-blockers is likely beneficial in heart failure with reduced ejection fraction, but the effect with ivabradine is relatively meager and can probably be achieved with better doses of β-blockers. Massie asked “Is a higher heart rate the mechanism for the poor outcome of heart failure patients, or is it a marker of people who are at higher risk? Isn‘t it likely that the mechanism for the consistently better results in the β-blocker trials is due to effects beyond heart-rate lowering?”

Finally, “Why were patients in SHIFT on such low doses of β-blocker? Only 26% was at target dose. Were these patients really unable to tolerate a higher dosage of β-blockers?” Massie is not sure about that, all the more because target doses of β-blockers were much higher in the large β-blocker trials CIBIS-2 (86%), MERIT-HF (90%), and COPERNICUS (74%). Also, from real life practice he knows of, higher doses of β-blocker seem possible. Massie showed, for the US Veterans Affairs hospital system, that by 2010, 75% of the patients with HFREF were receiving target doses of evidence-based β-blockers. Massie suggested that mechanistic trials should be carried out, to try and better characterize the effects of ivabradine on cardiac function and remodeling; mechanisms generally seen as important elements in the benefit of heart failure drugs, including β-blockers.

Böhm is not convinced that the US Veterans Affairs hospital system is representative of every day clinical practice. “The reality of everyday practice is still very low doses and low frequency of β-blocker prescription.” Chronic obstructive pulmonary disease is still very often a reason for nonprescription of β-blockers to patients who have an indication, irrespective of existing evidence of high-tolerance of especially cardio-selective β-blockers [5–7]. Also in the SHIFT study, concurrent chronic obstructive pulmonary disease was the main reason (37%) for not giving β-blockers to 344 patients in the trial. Other reasons included levels of hypotension (17%), asthma (10%), decompensation (7%) and fatigue (5%). “Such patients would generally not have been included in β-blocker trials,” Böhm added. According to Böhm, we have to realize that the SHIFT population differed from those in β-blocker trials. In the more contemporary SHIFT, approximately 90% of patients in both randomized groups received β-blockers, 91% received ACE inhibitors or ARBs, 83% received diuretics, 60% received aldosterone antagonists and 22% received cardiac glycosides. Therefore, as a whole the background therapy was much more intense than in β-blocker trials. This is especially true for aldosterone antagonists. Therefore, a lower risk reduction could be expected for ivabradine on top of all drugs commonly used, than with β-blockers, in addition to the heart failure drugs common in the 1990s. Bohm also stressed that β-blocker therapy was not held back from patients, intentionally or not, in order to give ivabradine in SHIFT. β-blocker use remained stable from start to finish (90% of all participants). The final argument of Böhm is that, although some nonsignificant, all secondary outcomes (Table 1) pointed in the same beneficial direction for ivabradine, and that also in patients who used >50% of target β-blocker dose, the benefits mirrored those of in the trial population as a whole, although, nonsignificant (nonsignificant, because it is a subgroup analysis).

“Heart rate reduction beyond what is achieved with β-blockers can further improve heart failure outcomes”. Böhm thinks that heart rate, in part, is insensitive to β-blocker treatment, and he speculates that there is a component of heart rate that is apparently independent of β-blocker treatment. Moreover, “the ‘only‘ determinant of poor prognosis on β-blocker therapy is still heart rate, and therefore, it makes sense to further reduce heart rate.”

The telemedical intervention studies TEHAF & Telemedical Interventional Monitoring in Heart Failure

Two multicenter studies were presented; the Telemonitoring in Heart Failure (TEHAF) study (The Netherlands) and the Telemedical Interventional Monitoring in Heart Failure (TIM-HF) study (Germany).

The TEHAF study included 382 patients (197 intervention and 185 usual care). Weight, blood pressure, but also pulmonary peak flow and blood glucose levels were monitored. The power calculation was rather optimistic with an expected reduction of 50% in heart failure admissions during 1 year follow up. The primary end point was time to first admission for heart failure. The mean age was 73 years, 59% were male, mean duration of heart failure was 1.25 years, and the mean LVEF was 36% (39% had a LVEF > 45%). Patients were classified as NYHA II (57%) or III/IV (43%) at baseline. Heart failure drug use was as follows; 87% diuretics, 58% ACE inhibitors, 33% ARBs, and 82% β-blockers. The primary end point, time to first admission for heart failure did not differ significantly between the two treatment groups with a HR 0.65 (95% CI: 0.35–1.18; p = 0.151). Of the secondary end points, the number of contacts with a heart failure nurse was significantly lower in the intervention group (267 vs 321; p < 0.01).

This study seems to be underpowered because of optimistic power calculations, low inclusion numbers and a relatively short follow-up period. As a result, a relative reduction of 35% in the time to first admission for heart failure with the intervention was not significant.

The TIM-HF study was also a randomized controlled trial. Within a large study, costing €16.2 million (Federal grant €7.15 million) and lasting from 2005–2011, the clinical trial (TIM-HF) was incorporated during 2008–2010. For this study, a remote patient monitoring system, including mobile sensor platform and electronic patient records was developed for the telemedical monitoring of heart failure patients. In total, 710 patients were randomized; 354 (49,9%) remote patient monitoring + guideline based therapy (RTM), and 356 (50,1%) guideline based therapy (control group, care as usual). Main inclusion criteria: NYHA class II/III, LVEF ≤35%, decompensation due to heart failure in the prior 24 months or LVEF ≤ 25%, considered optimally treated (i.e., ACEi/ARB, β-blocker and diuretic). Main exclusion criteria included hospitalization for worsening heart failure within previous 7 days and planned CRT implantation.

Patients were considered compliant to telemedical care when they used ≥70% of possible daily transfers (excluding hospitalizations) and when no “break” >30 days occurred. This was achieved by 287 (81%) of 354 RTM patients. The completeness of follow-up of the study was very good at 99.7%. A clinical end point committee decided on all unplanned hospitalizations using predefined criteria. The mean age was 67 years, 81% were male, 45% had an ICD, 50% were in NYHA class II, and another 50% in NYHA class III, and the mean LVEF was 27% (standard deviation 6). Heart failure medication use was high, with 96% on ACE inhibitor or ARB, 92% on β-blocker, 94% on diuretics and 64% on aldosterone antagonists. Approximately one out of five of the participants lived alone and the mean duration of heart failure was 6.8 (standard deviation 6.5) years. The primary end point was all-cause mortality. The results of the study were neutral. The HR of telemedical care for all-cause mortality was 0.97 (95% CI: 0.67–1.41) during 27 months of follow-up. Of the secondary end points, the physical functioning score with the SF-36 was significantly better for the complete period of 2 years (p < 0.05), however, it was overlapping with the care as usual group at the end of the study. Within the predefined subgroup of patients with a LVEF <2 5%, without depression and prior hospitalization for heart failure, the days lost due to death or heart failure hospitalizations were less in the telemedical care group with a HR of 0.48 (95% CI: 0.24–0.95). The presenter concluded that telemedicine seems not to be useful for all heart failure patients, but it could be a ‘bridge to stability‘ for unstable heart failure patients.

Both telemedical intervention studies could be interpreted as ‘neutral‘, although, with a tendency to better outcomes with the intervention (in certain subgroups). This finding is in contrast with a recent meta-analysis by Inglis et al. on telemedical intervention studies, published in 2010 [8], and in line with the largest telemedical study so far, the TELE-HF with 1653 patients included, also published in 2010 [8,9]. In the meta-analysis of Inglis et al. 11 studies (2710 participants) evaluated telemonitoring, resulting in a reduction of all-cause mortality (releative risk 0.66, 95% CI: 0.54–0.81; p < 0.0001) and heart failure related hospitalizations (releative risk 0.79, 95% CI: 0.67–0.94; p = 0.008) as compared with care as usual. The primary end point in the TELE-HF study was readmission for any reason or death from any cause within 180 days after enrollment. The telemonitoring group and the usual-care group did not differ significantly with respect to the primary end point, which occurred in 52.3 and 51.5% of patients (p = 0.75), respectively. Death occurred in 11.1% of the telemonitoring group and 11.4% of the usual-care group (difference, -0.2%; p = 0.88). Readmissions for any reason tended to be higher in the intervention group (49.3 vs 47.4%; difference 1.9%; p = 0.45).

Comparing telemedical intervention studies and meta-analyses of these studies are hampered by heterogeneity in the study designs, baseline characteristics of participants, and the methods applied. The discussion is ongoing on which elements are key for telemonitoring and which patients profit the most.

Table 1. Systolic Heart Failure Treatment with the If Inhibitor Ivabradine Trial (SHIFT): effect on primary and major secondary end points, ivabradine versus placebo.

Outcomes	Ivabradine n = 3241 (%)	Placebo n = 3264 (%)	HR (95% CI)	p-value
Primary end point
CV death or hospitalization for worsening heart failure	24	29	0.82 (0.75–0.90)	<0.0001
Secondary end points
All-cause death	16	17	0.90 (0.80–1.02)	0.092
CV mortality	14	15	0.91 (0.80–1.03)	0.128
Death from heart failure	3	5	0.74 (0.58–0.94)	0.014
All-cause hospital admission	38	42	0.89 (0.82–0.96)	0.003
Any cardiovascular hospital admission	30	34	0.85 (0.78–0.92)	0.0002
Hospitalization for worsening heart failure	16	21	0.74 (0.66–0.83)	<0.0001
CV death, HF hospitalization, or admission for nonfatal MI	25	30	0.82 (0.74–0.89)	<0.0001

CV: Cardiovascular; HF: Heart failure; HR: Hazard ratio; MI: Myocardial infarction.

Table 2. Systolic Heart Failure Treatment with the If Inhibitor Ivabradine Trial: hazard ratio for cardiovascular death or heart failure hospitalization by quintiles of increasing baseline heart rate, relative to lowest quintile.

Baseline heart rate quintiles bpm	Hazard ratio (95% CI)	p-value
70 to <72	1.00	–
72 to <75	1.15 (0.88–1.48)	0.308
75 to <80	1.33 (1.03–1.70)	0.027
80 to <87	1.80 (1.40–2.31)	<0.0001
>87	2.34 (1.84–2.98)	<0.0001

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.