246Effect of Ivabradine on Mortality in Patients with heart failure and a reduced left ventricular ejection fraction not receiving a beta-blocker: an analysis from SHIFT

Abstract

Background: Beta-blockers improve prognosis in patients with heart failure and a reduced left ventricular ejection fraction (HFrEF) in sinus rhythm. Whether blockade of adrenergic receptors is an essential mechanism of benefit or merely one method by which to reduce heart rate is uncertain. Ivabradine also reduces heart rate but does not block adrenergic receptors. Adding ivabradine to beta-blockers reduces heart failure hospitalizations but not the risk of sudden death. However, there are few data on the effect of ivabradine on outcome in the absence of a beta-blocker. Methods and Results: A post-hoc analysis of the SHIFT data was conducted investigating the effect of Ivabradine on all-cause and cardiovascular mortality amongst the subgroup of patients who were not prescribed a beta-blocker. Of 6505 patients valid for this analysis, 685 (10.5%) were not taking a beta-blocker of whom 398 (58.1%) were in NYHA class III or IV and 476 (69.5%) had coronary artery disease. Their mean age was 64 (SD11) years and LVEF 29 (SD5)%. Patients not taking beta-blockers had a mean baseline heart rate of 84 (SD12bpm), which had dropped to 68 (SD11) bpm at 28 days in those assigned to ivabradine and to 81 (SD14bpm) in those assigned to placebo. Patients assigned to placebo who were not taking beta-blockers had a higher risk profile and higher mortality (27.3%) compared to those taking beta-blockers (15.7%; p=0.009). Overall, 552 patients assigned to placebo died compared to 503 assigned to ivabradine (unadjusted HR 0.91; 95% CI 0.80-1.02; p=0.11). Amongst patients taking beta-blockers, there were 459 deaths amongst those assigned to placebo compared to 432 assigned to ivabradine (unadjusted HR 0.94; 95% CI 0.83-1.08; p=0.38). Amongst those not taking beta-blockers there were 93 deaths in those assigned to placebo but only 71 in those assigned to ivabradine (unadjusted HR 0.70; 95% CI 0.52-0.96; p=0.026). A test for interaction was of borderline significance (p=0.089). For those not taking beta-blockers there was a similar trend for cardiovascular mortality (81 versus 63 deaths; unadjusted HR 0.72 (0.52-1.00); p=0. 050). Conclusion(s): In SHIFT, patients who were not prescribed beta-blockers had higher heart rates and higher mortality than those prescribed beta-blockers and tended to have a greater benefit from ivabradine. This retrospective post-hoc analysis is consistent with the hypothesis that patients with HFrEF in sinus rhythm should receive a beta-blocker whenever possible but, when this is not the case, that ivabradine may reduce mortality in the absence of a beta-blocker. The strikingly similar point-estimate for the effect of beta-blockers and ivabradine ?monotherapy? compared to placebo on all-cause mortality for patients with HFrEF in sinus rhythm should be explored further.