Article Text

other Versions

Prevention of acute ischaemic events: an additional benefit of β-blockade with nebivolol in elderly patients with heart failure
  1. Giuseppe Ambrosio1,
  2. Marcus D Flather2,
  3. Michael Böhm3,
  4. Alain Cohen-Solal4,
  5. Adriano Murrone1,
  6. Flavio Mascagni5,
  7. Giulio Spinucci1,
  8. Maria Giovanna Conti1,
  9. Dirk J van Veldhuisen6,
  10. Luigi Tavazzi7,
  11. Andrew J S Coats8
  1. 1Division of Cardiology, University of Perugia School of Medicine, Perugia, Italy
  2. 2Clinical Trials and Investigation Unit, Royal Brompton and Harefield NHS Trust, London, UK
  3. 3Klinik für Innere Medizin III, Universität des Saarlandes, Homburg/Saar, Germany
  4. 4INSERM U942; Faculté Paris 7 Paris Diderot; Hopital Lariboisiere, Paris, France
  5. 5Data Management & Biometry Unit, CAST and TRIAL, Pozzo della Chiana, Italy
  6. 6Department of Cardiology, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands
  7. 7Department of Cardiology, IRCCS Policlinico San Matteo, Pavia, Italy
  8. 8Faculty of Medicine, University of Sydney, Australia
  1. Correspondence to Professor Giuseppe Ambrosio, Cardiology. University of Perugia School of Medicine, Ospedale Silvestrini, Via S. Andrea delle Fratte, 06156 Perugia, Italy; giuseppe.ambrosio{at}


Objectives This subanalysis of the Study of the Effects of Nebivolol Intervention on Outcomes and Hospitalisation in Seniors with Heart Failure (SENIORS) investigates whether treatment with nebivolol, a β-blocker with nitric oxide-releasing properties, can provide additional benefits besides its effects on heart failure (HF), by reducing cardiac ischaemic events in patients with HF of ischaemic aetiology.

Design A double-blind, randomised, placebo-controlled, multicentre trial of nebivolol in 2128 elderly patients.

Patients and Interventions For this analysis, data were extracted for 2128 elderly (≥70 years) HF patients in whom coronary artery disease (CAD) was the underlying aetiology (68.2%; 717 placebo-treated patients and 735 assigned to nebivolol).

Main Outcome Measures The main endpoint was the composite of cardiac ischaemic events at 2 year follow-up: death/hospitalisation for myocardial infarction, unstable angina or sudden death, as originally identified in the case report form.

Results At follow-up, nebivolol treatment was associated with a one-third reduction in the risk of ischaemic events, the composite endpoint occurring in 15.9% of placebo and 10.7% of nebivolol-treated patients (HR 0.68; 95% CI 0.51 to 0.90; p=0.008). This effect was independent of age, gender and ejection fraction. No difference in this composite endpoint was observed in the subgroup of patients of non-ischaemic aetiology.

Conclusions Nebivolol was effective in reducing cardiac ischaemic events in patients with HF of ischaemic aetiology. The prevention of ischaemic events can be an additional beneficial effect of β-blockade in HF patients with underlying CAD.

  • Acute coronary syndrome
  • β-blocker
  • coronary artery disease
  • heart failure
  • ischaemic events
  • nebivolol
  • nitric oxide

Statistics from

Heart failure (HF) is a common syndrome with an increased risk of death and disability. The prevalence of HF increases with age.1 The progression of HF has been shown to be related to increased activation of the renin–angiotensin–aldosterone and adrenergic nervous systems. The ability of β-blockers to counteract these effects provides a rationale for the choice of these drugs as a therapeutic approach in HF.2 Indeed, over the past 20 years large randomised controlled trials and meta-analyses of randomised trials have convincingly demonstrated the benefits on HF mortality and morbidity provided by β-blocker administration.3–8

β-Blockers are the mainstay therapy of chronic stable angina;9 furthermore, they are also effective in the secondary prevention of death and ischaemic events after myocardial infarction (MI).10–13 As coronary artery disease (CAD) is the primary cause of HF,14 it would be important to test whether, in the setting of HF, chronic β-blockade also beneficially affects ischaemic events, and whether this contributes to the efficacy of this treatment strategy. However, data in this respect are scanty, as previous clinical trials have not prospectively examined the effect of β-blockers on outcome in the specific subpopulation of HF patients with an ischaemic aetiology.

Among various guideline-recommended β-blockers, nebivolol is a β-1-selective blocker, with vasodilating properties that are attributed to nitric oxide (NO) modulation.15 In the Study of the Effects of Nebivolol Intervention on Outcomes and Hospitalisation in Seniors with Heart Failure (SENIORS) trial nebivolol administered on top of standard therapy to elderly (≥70 years old) HF patients significantly reduced the composite risk of all-cause mortality and cardiovascular hospital admission;16 this benefit occurred regardless of baseline ejection fraction (EF).17 In the setting of HF and chronic ischaemic heart disease, enhanced NO activity might further contribute to the beneficial effects of nebivolol, as it may provide anti-ischaemic protection against ischaemic injury to the myocardium by coronary vasodilation and decreased platelet and leucocyte activation.18

In the present study, therefore, we took advantage of the SENIORS database to perform a subanalysis with the specific aim of investigating whether chronic β-blockade by nebivolol can reduce ischaemic events in patients with HF of ischaemic aetiology.


The methods and main findings from SENIORS have been published.16 SENIORS was a randomised, double-blind, multicentre trial comparing the effect of nebivolol with placebo on top of standard therapy in elderly patients with HF. The main inclusion criteria were: age 70 years or older; a clinical history of chronic HF with one of the following: hospital admission within the previous 12 months with a discharge diagnosis of congestive HF; or left ventricular ejection fraction (LVEF) of 35% or less. Study medication was titrated over a 4–16-week period from a starting dose of 1.25 mg a day to a target of 10 mg a day.


A total of 2128 patients (1067 in the nebivolol group and 1061 in the placebo group) was available for intention-to-treat analysis. For the present study, a post-hoc analysis was performed on the SENIORS database, identifying the subgroup of patients in whom HF was of ischaemic aetiology, as recorded in a prespecified field of the original case record form.

Primary outcome measure and study endpoint

In the original study, the primary outcome was the composite of all-cause mortality or hospital admission for a cardiovascular event.16 Deaths were subclassified by the Clinical Event Review Committee into cardiovascular, non-cardiovascular, or of unknown cause. Cardiovascular admissions included acute MI, unstable angina, worsening of HF, cardiogenic shock, cardiac arrhythmia, stroke, peripheral vascular disease, dissection/rupture of aortic aneurysm, surgery/invasive procedure, bleeding, transient ischaemic attack and thromboembolic/vascular episode. Hospital admissions were defined as admissions to hospital involving a stay of at least 24 h (excluding admissions planned before randomisation and admissions for study-related procedures). All other causes of hospital admission were classified as non-cardiovascular or unknown if relevant information was missing.

Sudden death was defined as any unexpected death that could not be attributed to any identifiable cause, including ‘documented arrhythmia’, occurring as: (1) witnessed instantaneous unexpected death without any preceding symptoms or; (2) witnessed cardiac death occurring within 24 h after the onset of chest pain, syncope, acute pulmonary oedema or cardiogenic shock; or (3) non-witnessed unexpected death, if other causes of death could be excluded with reasonable certainty.

For the purpose of the current study, the primary endpoint was a composite of the following cardiac ischaemic events: hospitalisation/death for: acute MI, unstable angina or sudden death. Secondary endpoints were a composite of hospitalisation/deaths for acute MI or unstable angina and sudden death. The occurrence of these events (time to first event) was taken from the original database as adjudicated by the Clinical Event Review Committee. Ischaemic outcome was then further stratified by major clinical characteristics (gender, age, EF and diabetes), as prespecified in the original SENIORS study.16

Statistical methods

Descriptive statistics are presented as mean±SD or median and IQR (as appropriate) for continuous variables and the number of non-missing observations or as a percentage for categorical variables. Statistical analysis was carried out using the intention-to-treat principle according to a plan drawn up before the outcome data were available. The χ2 test and analysis of variance were used to assess the homogeneity of baseline characteristics between treatment groups. Outcomes were analysed using a Cox proportional hazards model with randomised treatment as the major covariate adjusted for baseline age, gender and EF according to protocol, to obtain the HR. For each HR, the 95% CI was computed. The outcome of specific subgroups is presented as Forest plots; in that case, if the 95% CI did not cover the 1.0 value, the difference between the two groups of treatment was considered statistically significant, corresponding to a two-sided test with α=5%. The effect of nebivolol on major clinical subgroups (gender, age, EF and diabetes) was assessed by tests of interaction in Cox proportional hazard regression analyses adjusted for age, gender and EF. A two-sided p value of less than 0.05 was considered statistically significant. Statistical analysis was performed using SAS software version 9.1.


Study patients

Of the 2128 patients enrolled in the SENIORS study, 1452 (68.2%) had a previous history of CAD documented on the case record form and were included in the analysis for the present study. Patient numbers were similar in the placebo (n=717) and nebivolol-treated groups (n=735). Demographic and baseline clinical characteristics for these patients are shown in table 1. Apart from a higher proportion of female patients within the nebivolol group, there was no difference between baseline characteristics for the two study groups. The majority of patients (∼90%) were distributed between class II and III of the New York Heart Association classification. Half of the patients within either the placebo or nebivolol-treated groups had had a previous MI, hypertension or hyperlipidaemia. Background medication was also similar between both groups, and ACE inhibitors/angiotensin II antagonists were used by more than 90% of patients.

Table 1

Baseline characteristics of patients with CAD

Effect of nebivolol on study endpoints in CAD patients

Primary endpoint

After a median follow-up of 20.1 months (IQR 13.2–30.5 months) in CAD patients receiving placebo and 20.5 months (IQR 13.7–30.7 months) in nebivolol-treated CAD patients, the proportion of patients reaching the primary endpoint of composite ischaemic events was 15.9% in the placebo group and 10.7% in the nebivolol group, representing a one-third proportional reduction (unadjusted HR 0.66, 95% CI 0.50 to 0.88; p=0.0051). This difference remained significant after adjusting for gender, age and LVEF (5.2% absolute reduction; HR 0.68, 95% CI 0.51 to 0.9; p=0.008; table 2).

Table 2

Outcomes (time to first event) of patients with CAD (adjusted)

The time course of composite ischaemic events is presented in figure 1, and shows that the event curves for nebivolol-treated patients compared with placebo had already separated before 6 months, and continued to diverge further for the duration of follow-up.

Figure 1

Kaplan–Meier plot of primary endpoint. The primary endpoint (composite of all ischaemic events as described in the Methods section) for SENIORS patients of ischaemic aetiology (SENIORS–coronary artery disease), for nebivolol versus placebo; HR 0.68, 95% CI 0.51 to 0.9; p=0.008.

Secondary endpoints

Nebivolol treatment was associated with a significant reduction in sudden death during follow-up (HR 0.62, 95% CI 0.41 to 0.92; p=0.017; table 2; figure 2 upper panel). Hospitalisation/death for acute MI or unstable angina occurred in 8.4% of placebo-treated patients during follow-up, and in 6.1% of patients receiving nebivolol (figure 2, lower panel, table 2; p=0.12).

Figure 2

Kaplan–Meier plot of secondary endpoints. The effect of nebivolol versus placebo during follow-up was determined with respect to the occurrence of sudden death (A) HR 0.62, 95% CI 0.41 to 0.92; p=0.017 and hospitalisation/deaths for acute myocardial infarction or unstable angina (B) HR 0.74, 95% CI 0.50 to 1.09; p=0.12.

Outcome in specific CAD patient subgroups

Figure 3 shows the results for the primary endpoint when patients were stratified by major clinical variables (gender, age, EF, diabetes), as prespecified in the original SENIORS trial. The beneficial effect of nebivolol on ischaemic events in this HF population was reproduced across all subgroups. Furthermore, the magnitude of benefit was similar, regardless of the clinical variable used for grouping. In particular, HR for men was 0.68 (95% CI 0.49 to 0.96), and for women 0.64 (95% CI 0.36 to 1.13) (for interaction test p=0.86; figure 3). The HR for patients aged less than the median of 75.2 years was 0.62 (95% CI 0.41 to 0.93) and it was 0.73 (95% CI 0.49 to 1.1) in those aged over 75.2 years (for interaction test p=0.42; figure 3). The HR for patients with an LVEF of 35% or less and greater than 35% were 0.65 (95% CI 0.46 to 0.92) and 0.71 (95% CI 0.43 to 1.19), respectively (for interaction test p=0.76; figure 3). The HR for patients with diabetes was 0.74 (95% CI 0.44 to 1.26) and 0.65 for patients without diabetes (95% CI 0.46 to 0.92) (for interaction test p=0.77; figure 3).

Figure 3

Primary outcome by SENIORS–coronary artery disease (CAD) prespecified subgroups (HR with 95% CI). *Annual rate as number of events per 100 patient-years of follow-up at risk. **p Value for interaction: age and ejection fraction considered as continuous variables. LVEF, left ventricular ejection fraction.

Additional analyses

To characterise fully the effect of nebivolol compared with placebo in our study, we ran two additional analyses.

Effect of nebivolol on original primary outcome in CAD patients

To verify whether the initial findings of the SENIORS trial would also hold true in the cohort of patients with ischaemic aetiology in the present study, the effect of nebivolol compared with placebo was also assessed with respect to the original endpoint of the overall SENIORS trial (ie, all-cause mortality or cardiovascular hospitalisation). Also, within this ischaemic cohort, the proportion of patients who had an event in the nebivolol group was lower compared with the placebo group (30.3% vs 34.5%; HR 0.84, 95% CI 0.7 to 1; p=0.055); nebivolol-treated patients with HF of ischaemic aetiology achieved the same level of benefit that was previously observed in the entire cohort,16 with an absolute risk reduction of 4.2%, (RR reduction of 12.2%), although possibly due to the smaller sample size, this difference failed to reach statistical significance (p=0.055).

Effect of nebivolol on ischaemic events in non-CAD patients

Conversely, in another analysis, we evaluated whether the administration of nebivolol was associated with a different rate of ischaemic events in the subgroup of patients in whom the underlying aetiology was not ischaemic. This subgroup comprised 676 patients, of whom 344 were randomly assigned to placebo and 332 to nebivolol. As expected, this subgroup experienced fewer ischaemic events compared with the group with ischaemic aetiology: by the end of follow-up, the composite primary ischaemic endpoint was similar in both groups, ie, 10.8% of patients receiving placebo and 9.6% of patients randomly assigned to nebivolol (HR 0.86, 95% CI 0.53 to 1.39; p=0.54).


The main finding of this analysis of the SENIORS trial shows that nebivolol treatment in elderly patients with HF of ischaemic origin was associated with a marked reduction in cardiac ischaemic events at follow-up, which remained significant after adjusting for possible confounders. Furthermore, no differences were observed when patients were categorised according to age, gender, reduced EF, or the presence of diabetes, as the effect of nebivolol was similar across all of these subgroups.

β-Blockers, including nebivolol, are the recommended therapy for patients with HF and decreased EF.3–8 Furthermore, the results of SENIORS demonstrated that nebivolol can also beneficially affect outcome in HF patients with preserved EF.17 In the setting of HF, β-blockers blunt the activation of the adrenergic system and indirectly also of the renin–angiotensin system, which are known to play a key role in the progression of HF and adverse cardiac remodelling.3 However, in addition to these effects, β-blockers are endowed with specific properties that are of value in the management of patients with CAD.9 This is a potentially important feature, given the fact that ischaemic heart disease is the predominant aetiology of patients with HF in developed countries.14

Large randomised clinical trials have demonstrated improved outcome in HF patients treated with β-blockers on top of standard therapy with diuretics and ACE inhibitors. Among these, the CAPRICORN19 and the Australia–New Zealand Heart Failure20 trials enrolled only patients with a known history of CAD, while in the COPERNICUS,6 COMET,7 21 BEST,8 US–CARVEDILOL22 and MERIT–HF4 trials, ischaemic heart disease was the underlying aetiology in the majority of patients. Furthermore, post-hoc analysis of MERIT–HF demonstrated a 40% reduction in total mortality in the post-MI subgroup of patients.23 It is well established that β-blockers consistently improve survival and reduce rehospitalisation for HF in patients with chronic ischaemic heart disease. Much less investigated, however, is whether the occurrence of ischaemic events, which are the likely sequelae of underlying CAD, is also reduced in this large group of HF patients.

Earlier studies provided an indication that β-blockers could significantly reduce mortality and reinfarction in patients surviving acute MI.10–12 More recently, a 41% reduction in non-fatal MI was seen in the CAPRICORN trial, which enrolled only post-MI patients.19 However, with respect to the effects on cardiac ischaemic events in HF patients outside the post-MI setting, previous studies with β-blockers only offer limited information. In particular, no data on fatal MI are reported for MERIT–HF,4 CIBIS–II,5 and COPERNICUS;6 whereas in COMET,7 21 BEST8 and the US–CARVEDILOL22 trials, no significant decrease in MI-related deaths was seen with treatment. It should be noted, however, that these studies did not specifically focus on patients with an ischaemic aetiology. As for non-fatal MI, data are also largely inconclusive: a significant reduction in non-fatal MI was retrospectively found in bucindolol-treated,8 but not in bisoprolol, carvedilol or metoprolol-treated patients.5 6 23 24 Finally, sudden death in HF patients was found to be significantly reduced in MERIT–HF,4 CIBIS–II5 and the Australia–New Zealand Heart Failure20 trials, but not in other trials;19 22 25 however, also with respect to reporting rates of sudden death, data did not specifically refer to patients with HF of ischaemic aetiology. In this analysis, the reduction in sudden death was the major contributor to the reduction in the composite endpoint. We consider sudden death an ischaemic event because: (1) acute myocardial ischaemia is in fact the most common factor triggering fatal arrhythmias in HF patients;14 26 27 (2) the definition of sudden death as outlined in SENIORS encompasses patients who actually died up to 24 h after initial symptoms, and it is established that sudden death occurring several hours after the onset of symptoms is quite often related to arrhythmias arising in the setting of acute myocardial ischaemia or MI.26

This is thus the first assessment that specifically addresses cardiac ischaemic outcomes in this setting. In this respect, we report that a β-blocker may confer additional anti-ischaemic benefits in the large subgroup of HF patients with an underlying ischaemic aetiology. It is conceivable that this characteristic is also shared by other β-blockers, and that it has simply gone undetected because previous studies did not systematically look for this type of effect. At the same time, it is also possible that our findings are due to certain specific features of the original SENIORS trial, which may have made the anti-ischaemic effects of β-blockade more evident. In this respect, it should be noted that the duration of follow-up was longer in SENIORS than in previous trials. Therefore, it is possible that in SENIORS there has been a greater potential for ischaemic events to accrue over time. Another major difference with previous trials is that the average age of patients was over 75 years in the present study; older patients may be more susceptible to ischaemia-induced events.28

Whereas these factors may explain some of the differences between SENIORS and earlier trials, it should also be underlined that nebivolol is known to possess additional properties that go beyond its action as an antagonist of β-adrenergic receptors,15 29 and that may be relevant with respect to its anti-ischaemic action. These ancillary beneficial properties revolve around the capability of nebivolol to stimulate NO production secondary to increased activity (and overexpression) of NO synthase by activation of β-3-adrenergic receptors,30 and to prolong NO half-life by an antioxidant effect.31 The anti-ischaemic effects are well documented and include increased coronary dilation and coronary blood flow and decreased platelet and leucocyte activation.15 29 Therefore, enhanced release of NO by nebivolol (not shared by other β-blockers) may contribute to the improved anti-ischaemic profile of this drug.

In addition to these direct anti-ischaemic effects of vasodilation and inhibition of platelet activation, there may be one additional mechanism by which NO, and NO-potentiating agents (such as nebivolol) may induce cardiac protection against ischaemic events, namely ‘pharmacological preconditioning’.32 It is well known that brief ischaemic episodes can paradoxically precondition the heart (ie, make it more tolerant to subsequent major ischaemic insults).33 A fundamental step in the sequence of events eliciting ischaemic preconditioning is the activation of NO synthase, with its attendant production of NO.32 34 In experimental studies, a similar protective phenotype can be pharmacologically mimicked—in the absence of previous ‘preconditioning’ ischaemia—by pretreatment with NO donors,34 or with drugs that increase NO availability.32–34 In patients, treatment with NO donors significantly reduces myocyte necrosis following coronary angioplasty,35 and it is associated with evidence of less severe myocardial injury in the setting of acute coronary syndromes.36 In fact, the involvement of NO has been postulated as the ultimate mechanism of many infarct-sparing therapies.32 34


Although we analysed data that were already prespecified in the original trial and captured in the case record form, our findings are the result of a post-hoc analysis. Furthermore, ischaemic aetiology was not a specific requisite at entry, and patients were not randomly allocated to treatment, nor stratified, on the basis of that. Finally, our findings cannot be generalised to patients less than 70 years old. These data should thus be regarded as hypothesis-generating; prospective, randomised trials may ultimately establish the merits of this therapy. Also, although acute myocardial ischaemia accounts for most fatal arrhythmias in HF patients,14 26 27 non-ischaemic mechanisms of arrhythmogenicity (eg, anatomical/functional substrate, drugs, electrolytes, etc) may have also contributed to inducing sudden death.

In this study, gender was not strictly matched between the placebo and nebivolol-treated groups; however, subgroup analysis demonstrated that gender did not modify the effect of nebivolol. Indeed, subgroup analysis of prespecified major clinical categories (including EF and diabetes) shows that all subgroups benefited to a similar extent.


Ischaemic heart disease is the underlying aetiology in most cases of HF, and ischaemic events occur frequently in these patients. In this analysis of the SENIORS trial, cardiac ischaemic events at follow-up were significantly reduced in elderly patients with HF of ischaemic aetiology treated with nebivolol, a β1-receptor blocker and NO-releasing agent. Our findings support the hypothesis that β-blockers may also benefit HF patients with previous coronary disease through anti-ischaemic mechanisms, and provide further evidence of the importance of nebivolol in elderly patients with HF.


The authors would like to thank Dr Colin G Egan, who provided skillful editorial assistance.


View Abstract


  • Funding The original SENIORS trial was sponsored by Menarini. GA has received honoraria and speaker fees from Menarini. MDF has received research grants and honoraria from Menarini. MB has received speaker honoraria from Berlin Chemie and Menarini. DJV has received lecture fees from Menarini. AJSC and AC-S have received honoraria and speaker fees from Menarini.

  • Competing interests None.

  • Ethics approval This is a spin off of a multicentre trial that had been originally approved by the ethics committees for each participating site.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.