Background

High heart rate is associated with longevity in many animal species including men.1 It has been shown in various cardiovascular diseases that high resting heart rate is a known marker of cardiovascular outcomes in hypertension,2 atherosclerosis,3 myocardial infarction4 and heart failure.5 ,6 Among the different conditions the threshold from which risk is increased is different. More close correlations at low thresholds (≥70 bpm) to risk have been established in heart failure.5–8 One beat increase of baseline heart rate and five beats increase of resting heart rate is associated with an increase of cardiovascular death and heart failure hospitalisation of 3% and 16%, respectively.6 Therefore, it was tempting to speculate that heart rate reduction reverses risk, in particular in heart failure, where the association of heart rate and risk is very close. Furthermore, the pathophysiology is plausible for heart failure, because the force-frequency relationship is reversed in the failing heart in vitro9 ,10 and in vivo.11 Shortening of diastole reduces oxygen supply to the myocardium. High heart rates are associated with atherosclerosis12 ,13 potentially important in energy starvation of the failing heart.7 ,8 Finally, high heart rates are a reflection of neurohormonal activation and are associated with malignant arrhythmias.14–16 A reduction of events by heart rate reduction only, would qualify heart rate as a target of treatment and risk factor beyond representing just a surrogate or risk marker.

Intervention to reduce heart rate

Evidence for heart rate being indeed one of the crucial pathophysiological steps in progression of left ventricular dysfunction, has already been scrutinised by β blocker trials in heart failure.17–20 There is evidence that β blockers provide protection from heart failure related endpoints by reducing heart rate, although other mechanisms like direct inhibition of β adrenoceptors or vasodilatation for particular β blockers like carvedilol or nebivolol are possible.20 ,21 However, a meta-analysis in almost 20 000 patients from β blocker trials has shown that neither β blocker dose, the type of β blockers underlying the pathophysiology of heart failure is important, but only the amount of heart rate reduction is associated to a reduction of cardiovascular outcomes and even death.20 However to prove this concept of heart rate being a true target beyond representing only a surrogate or just a marker of risk, it was necessary to perform a trial with a drug exclusively reducing heart rate without any other known cardiovascular effect like vasodilatation or direct myocardial effects.

Ivabradine binds to non-selective cation I_f-channels in the sinus node and reduces heart rate without any other known cardiovascular effects.22 ,23 It provides potent antianginal effect indicating that in humans heart rate reduction is working to improve the energy consumption and supply in coronary artery disease.23 ,24 Furthermore, in patients after myocardial infarction with impaired left ventricular function, it failed to reduce the primary endpoint, although it reduced a new onset of non-fatal myocardial infarctions and the necessity for coronary revascularizations.25 This concept fits well with experimental observations that heart rate reduction reduces atherosclerosis26 and stroke size after stress in experimental animals27 due to an improvement of endothelial function.28 A reduction of the incidence of plaque ruptures might be important to reduce acute coronary events.29

Although associated with death, cardiovascular death and heart failure hospitalisations to this point, elevated heart rates appear only as a risk marker and not as a risk factor,30 (figure 1). However, in the systolic heart failure treatment with the If inhibitor ivabradine (SHIfT) trial performed on a population of 6.505 patients in chronic heart failure with reduced ejection fraction ≤35% at a heart rate >70 bpm in sinus rhythm, treatment with ivabradine with a target dose of 7.5 mg twice daily on top of standard contemporary medical treatment, significantly reduced the composite endpoint of cardiovascular death and heart failure hospitalisation by 18% resulting in a number needed to treat of 26 to prevent one event per year.31 This effect was obtained in a heart failure population already receiving heart rate reducing β blocker treatment in 89% of the cases and in whom 56% receiving at least 50% of the β blocker target dose was used as recommended by the Guidelines of the European Society of Cardiology.32 Because the intervention with selective I_f-current inhibition in the sinus node reduces events, this observation is in favour of the notion that heart rate is not only a surrogate and a risk marker, but also a risk factor.6 This finding is further substantiated by adjusting the beneficial effects of ivabradine to the number of reduced heart rates. The RR reduction with HR of 0.82 (0.75–0.9, p<0.0001) moves towards 1.0 (ie, 0.95 (0.85–1.06, p=0.352) after adjustment for the number of heart beats reduced after 28 days, that is, after uptitration of ivabradine.6 Furthermore, the reduction of events was closely associated to a reduction of heart beats6 ,33 and to low heart rate achieved with heart rates below 60 bpm providing the best protection after initiation of treatment.6 Therefore, the close correlation of baseline heart rate to risk, the close association of heart rate reduction and low heart rates achieved to risk reduction as well as the neutralisation of the beneficial effect after adjustment to the number of heart beats reduced, provides strong evidence that heart rate is not only a surrogate or marker but rather a true treatment target of heart failure.6

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Cardioprotective effects of heart rate reduction.

Predicting positive treatment effects, it appears important to judge heart rate attributable risk by determining the threshold heart rates from which risk is increased. In patients with impaired left ventricular function after myocardial infarction it was shown that the risk of cardiovascular death and heart failure in coronary artery disease related events is increased above 70 bpm.6 This was one rationale to perform the SHIfT trial in patients at elevated risk, that is, above 70 bpm in sinus rhythm.34 It turned out that the increase of risk was homogeneous from 70 bpm for the composite cardiovascular endpoint heart failure hospitalisation and cardiovascular death,6 while the increase of risk for the component cardiovascular death of the composite was only increased at heart rates above 75 bpm.6 This might have been the motivation of the European Medicines Agency to approve ivabradine for the treatment of patients with heart failure and systolic dysfunction receiving guideline based recommended background therapy at resting heart rate ≥75 bpm in sinus rhythm. It has to be pointed out that this decision was based on a non-prespecified subgroup and a retrospective analysis of the SHIfT trial. An in-depth analysis on this subpopulation of patients in SHIfT of a heart rate above 75 bpm has been published recently33 showing that the primary composite as well as all components thereof and all secondary endpoints were strongly and significantly reduced in the population of patients with heart rates ≥75 bpm, while these effects were less pronounced and more or less completely non-significant in patients at systolic heart rate ≥75 bpm.33 Again in both populations, risk reduction was associated to a number of heart beats reduced and the low heart rates achieved after uptitration of ivabradine on top of recommended contemporary treatments including β blockers.31 Therefore, the approval of the drug is now based on a non-prespecified subgroup, which is less strongly evidence-based than the data of the main SHIfT trial. Therefore, the new Guidelines of the European Society of Cardiology recommend the treatment of heart failure population in sinus rhythm ≥70 bpm,31 because this evidence is generated in a prospective randomised trial in a broad and well treated population of patients.6 ,31 Furthermore, following the approval of the European Medicines Agency, patients at heart rate <75 bpm and a reduction to lower more than 50 bpm on treatment with ivabradine would still benefit from such treatment concerning heart failure hospitalisations and are not covered by the present approval of the drug.33

Cardiac glycosides

It is interesting to observe that apparently heart rate reduction with other drugs might also be related to risk reduction and, therefore, the concept could apply to a broad spectrum of pharmacological interventions. A recent secondary analysis34 of the Digitalis Investigation Group trial with digoxin35 has shown that a similar reduction of the heart failure related endpoints cardiovascular death and heart failure hospitalisation by 15 (9%–21%) was also observed with digoxin,34 a trial in which digoxin presumably reduced heart rate, although heart rate was not directly captured.35 However, trials on the effects of withdrawal of digoxin have shown increase of events after withdrawal accompanied by an increase of heart rate of 7 bpm over 3 months.36 ,37 Therefore, one might speculate that heart rate is a general marker of risk, which might be a target of various pharmacological interventions including digoxin, β blockers and ivabradine. However, it is noteworthy that at the time the Digitalis Investigation Group trial was performed there were no contemporary treatments used, in particular no β blocker treatment of heart failure patients. Although, treatment with mineralocorticoid receptor antagonists was apparently high, although not precisely reported, the β blocker treatment was not approved or recommended at that time. Therefore, poor background therapy, in particular the low prevalence of β blocker treatment, makes the trial very sensitive for risk reduction by any intervention reducing heart rate. Therefore, it is open, if not unlikely, that in the presence of β blockade, Digitalis would have provided the same protection, although this analysis supports the role of heart rate as a modifiable risk factor.

Beta blockade and heart rate reduction with ivabradine

It is noteworthy that in SHIfT the risk reduction was only dependent on heart rate at baseline and heart rate reduction rather than on β blocker treatment or, in particular, on the dose of the β blockers applied.38 This is suggestive that heart rate is a true treatment target and further strengthens the observation from a meta-analysis in almost 20 000 patients that risk reduction of β blockers was only dependent on heart rate reduction achieved by β blockade rather than the type of β blockade, the dose of the β blockade, the underlying course of heart failure and many other potential confounders.20 It is important to note that there is no evidence that heart rate reduction with ivabradine can substitute adequately dosed β blocker therapy. SHIfT was designed to test ivabradine on top of recommended therapies including β blockade at the best tolerated dose. In the case report forms the β blocker had to be captured and all reasons for not uptitrating or even not applying a β blocker had to be justified by the investigators. Therefore, the trial in its design took care of potential underdosing of β blockers. In this respect it is noteworthy that SHIfT provided the best β blocker treatment intensity among all trials exclusive those who tested blockers in the treatment arms.

Summary and clinical perspective

In summary clinical and experimental evidence support the concept of heart rate being a true treatment target rather than only a surrogate in heart failure. This is supported by experimental studies in heart failure models, in models with atherosclerosis or stroke and in patients with heart failure. Further interventional studies in patients with hypertension, stable coronary artery disease and potentially also in primary prevention beyond observational studies in epidemiological cohorts could provide evidence that heart rate reduction is a general biological principal to reduce cardiovascular events and to prolong life in affected individuals. However, these trials are difficult to perform because the number of patients' years might be excessively high to observe sufficient event numbers. Up to now ivabradine has shown to reduce endpoints in chronic heart failure closely associated to heart rate achieved and absolute heart rate reduction in a population of patients in sinus rhythm ≥70 bpm for heart failure hospitalisations and ≥75 for total death, cardiovascular death and other vascular and heart failure related morbidity and mortality endpoints as well as improving left ventricular remodelling39 and quality of life.40 Therefore, the evaluation of cardiovascular risk by measuring resting heart rate becomes a new paradigm, which might be of similar importance as blood pressure, and provides a target for effective treatments with optimally dosed β blockers and ivabradine when heart rate remains above 70 bpm, potentially even in other conditions like heart failure.