Reported by Steve Stiles, theheart.org: May 22 2011
Like attorneys arguing before a jury, they both strengthened the case for perseverance in getting beta blockers to more HF patients at optimal dosage levels, took a snapshot of where cardiology stands in that quest, and provided a more penetrating look at—and deeper critique of—the randomized Systolic Heart Failure Treatment with the If Inhibitor Ivabradine Trial (SHIFT) [1] than could be aired when it was presented nine months ago at the main scientific sessions of the ESC.
In that trial, which randomized >6500 patients with systolic heart failure on standard medications to ivabradine (Procoralan, Servier) or placebo, those receiving the novel heart-rate-reducing agent showed a significant 18% drop in the composite rate of cardiovascular death or heart-failure hospitalization (p<0.0001) over an average of about two years, as reported then by heartwire.
"A valid therapeutic target"
HFA president and debate moderator Dr Piotr Ponikowski (Medical University, Fourth Military Hospital, Wroclaw, Poland) told heartwire: "I personally believe there is a lot of evidence [that] heart-rate reduction itself may be seen as a valid therapeutic target." If true, it provides a foothold for ivabradine, which works on an ion channel with pronounced expression in the sinoatrial node.
Ponikowski cautioned that as a former SHIFT point person at his center and researcher with Servier-sponsored trials of the drug, he "treats many patients with ivabradine." Still, the drug is only for patients optimally treated or intolerant of beta blockers. "If your patient is optimally treated with all the neurohormonal blockers, including beta blockers, and still the heart rate is about 70 [bpm], then there is a space to start with ivabradine, because it's a quite well-tolerated drug." He said patients should also be in sinus rhythm to receive it.
Dr Michael Böhm (Saarland University Hospital, Homburg/Saar, Germany), assigned to
make the case for ivabradine during the debate, observed for heartwire that the drug could potentially extend heart-rate-lowering drug therapy to heart-failure patients who—for whatever reason—are not on optimal beta blockers.
"We have patients who don't want to take [beta blockers], refuse to take them, and also those with side effects like hypotension or bradycardia that don't allow them to use them," he said. "And so therefore we end up with a rather low level of beta blockade in the general population. These patients might benefit on ivabradine. But SHIFT was designed to have it added on; it's not a substitute in patients who tolerate [beta blockers]."
. . But back to the show
The debate's real issue, according to Böhm's opponent, HFSA president Dr Barry M Massie (University of California and San Francisco VA Medical Center), "is whether ivabradine provides additional benefit in patients who are on optimal medical therapy." Patients in SHIFT, he advanced, were not.
The significantly reduced hazard for the primary end point in SHIFT among patients on ivabradine "was driven primarily by the reduction in heart-failure hospitalization, accounting for 514 of 793 events." The risk reduction was only 10% (p=0.092) for all-cause mortality and 9% (p=0.13) for cardiovascular mortality.
In contrast, the reduction was 26% (p<0.0001) for heart-failure hospitalization, 11% for all-cause hospitalization (p=0.003), and 18% for the composite of CV death/HF hospitalization/hospitalization for nonfatal MI (p<0.0001).
Isn't it likely that the mechanism for the consistently better results in the beta-blocker trials is due to effects beyond heart-rate lowering? But heart rate has long been associated with mortality and other clinical outcomes in these patients, ever since the dawn of beta-blocker therapy for heart failure, he noted. In the 1999 Cardiac Insufficiency Bisoprolol Study (CIBIS-2) and the more recent Metoprolol-CR/XL Randomized Intervention Trial in Congestive Heart Failure (MERIT-HF) and Carvedilol Prospective Randomized Cumulative Survival Study (COPERNICUS), all-cause mortality fell by at least a third in patients given beta blockers. The benefits were all highly statistically significant, according to Massie, in contrast with the drop in all-cause mortality in SHIFT that fell short of significance.
"All-cause mortality results [have been] consistently excellent in the beta-blocker trials; but in SHIFT, [ivabradine was] not very powerful."
He pointed to a 2009 meta-analysis of beta-blocker trials by McAlister and colleagues [2]: for every 5-bpm reduction in heart rate with beta blockers, mortality fell 18%.
After SHIFT, there remain some unresolved questions, Massie said. Reducing heart rate beyond what is currently accomplished with beta blockers is likely beneficial in heart failure. "I just think the effect is relatively meager and can be achieved probably better with other therapies or better doses of beta blockers," he said.
Background therapy . . . was much more intense than in any other trial before. . . . So, the risk reduction seen was much lower. But Massie continued, "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? We don't really know for sure. If it is an important mechanism, does it matter how it's lowered? Isn't it likely that the mechanism for the consistently better results in the beta-blocker trials is due to effects beyond heart-rate lowering?"
Finally, "Why were the patients in SHIFT on such low doses of beta blockers?" Target doses were achieved in 86%, 90%, and 74% of patients in the CIBIS-2, MERIT-HF, and COPERNICUS trials, respectively. "In SHIFT it was only 26%. . . . Were these patients really unable to tolerate a higher dosage of beta blockers? I'm not sure. This is what we've done in the VA system."
Change in beta-blocker use in heart failure in the US Veterans Affairs hospital system, 2007-2010 (shown by Massie at Heart Failure Congress 2011)
| Parameter |
Before 2007 (%) |
By 2010 (%) |
| On beta blockers |
40 |
90 |
| On evidence-based beta blockers |
25 |
85 |
| On >80% target beta-blocker dose |
60 |
85 |
| On target dose of evidence-based beta blockers* |
40 |
75 |
*Carvedilol or metoprolol
"What do we need to know before we shift to ivabradine? We have to improve the use of beta blockers. These are the single most effective pharmacological therapy for heart failure. The benefits reported in SHIFT are not comparable to those seen in beta-blocker trials. I think we should conduct a head-to-head trial of ivabradine vs placebo in patients taking reasonable doses [at least 75% of target dose] of evidence-based beta blockers to determine whether additional benefit can be achieved. And I think it would be nice to do some mechanistic trials and better characterize the effects of ivabradine on cardiac function and remodeling, since we know these are probably very important elements of the benefit of beta-blocker therapy."
Such comparisons "artificial"
SHIFT, Böhm said, represented something of a "proof-of-concept" study for the question of whether there is a causal relationship between heart rate and outcomes in heart failure or whether heart rate is simply a risk marker. Ivabradine, he said, "only reduces heart rate without any other known effect." In the trial, it reduced the average heart rate by 11 bpm compared with placebo.
Although the trial was not powered to find statistical significance in any of the components of the composite end point, their changes were "all in the same direction," adding confidence to the primary finding of a 26% risk reduction.
Moreover, it is "artificial" to judge the effect of heart-rate reduction in heart failure on the basis of beta-blocker trials, Böhm insisted when making the contrary argument: in clinical practice, prescription rates vary for beta blockers. "Apart from clinical trials, the reality is quite different [in that] there are in particular very low doses and very low frequency of beta-blocker prescription."
And the SHIFT patient population differed from those in beta-blocker trials, he said, as did the era of heart-failure therapy: in the more contemporary SHIFT, about 90% of patients in both randomization groups received beta blockers; 91% ACE inhibitors or angiotensin receptor blockers, 93% diuretics, and 60% aldosterone antagonists.
"So background therapy as a whole was much more intense than in any other trial before, and in particular, concerning aldosterone antagonists, it was much more intense than in the beta-blocker trials. So, the risk [reduction] seen was much lower."
It is true that only 26% of patients were at target beta-blocker dose and only 56% were at >50% of target dose, Böhm said. But all clinicians in the trial had to document their reasons for not giving beta blockers to patients: "There was a special effort to make sure that, indeed, these patients were on optimal medical treatment."
Their reasons for not giving beta blockers to 344 patients in the trial: chronic obstructive pulmonary disease (37%), hypotension (17%), asthma (10%), decompensation (7%), and fatigue (5%). Such patients would generally not have been included in beta-blocker trials "because they all had some contraindication against beta blockers."
Critics of the trial contend that, intentionally or not, its clinicians somehow held back on beta-blocker use "in order to give ivabradine, but this was not the case," according to Böhm. Beta-blocker use remained stable from start to finish: 88% to 90% at six, 12, 18, and 24 months.
Despite all that beta-blocker use, "still the effect [of ivabradine] on heart-failure hospitalization remained robust." Even in the >3100 patients who were at >50% of target beta-blocker dose, the benefits mirrored those in the trial population as a whole, with a 10% primary-end-point reduction ("nonsignificant, because it's a subgroup analysis"), but no reduction in all-cause mortality or CV death; a 10% reduction in all-cause hospitalization (a trend); and 12% and 19% reductions in CV (p=0.046) and HF hospitalization (p=0.021), respectively.
Heart rate: Culprit or bystander?
Böhm pointed to analyses of three European registries of >7000 heart-failure patients combined in which the prevalence of heart rate of at least 70 bpm—a SHIFT entry requirement—was 53% to 55%. About one-third of each registry had a heart rate >75 bpm and in about one-fifth it was >80 bpm. Yet the patients, regardless of baseline heart rate, were treated with beta blockers to the same extent, "indicating that heart rate, in part, is insensitive to beta-blocker treatment."
In SHIFT, he pointed out, the higher the baseline heart rate, the more heart rate dropped off with ivabradine; it fell by a mean 16 bpm for patients at 80 bpm to <87 bpm at baseline and by a mean 22 bpm for those who had been at >87 bpm. Moreover, the primary-end-point reduction—which reached significance with ivabradine in the overall population—was driven entirely by benefit among patients with a baseline heart rate of at least 80 bpm.
Now, beta blockers also work, at least in part, by heart-rate reduction, which has been shown to be significantly predictive of benefit in beta-blocker trials, he reminded the audience. Also in those trials, he said, "the higher the [baseline] heart rate, the more the benefit, just like with ivabradine."
So was the benefit in SHIFT due to heart-rate reduction "or something else from the treatment?" There was evidence in the trial that the benefit in those who took ivabradine was not independent of the drug's effect on heart rate, "indicating that, indeed, heart-rate [reduction] is the mechanism [of benefit] here: nothing beyond heart rate."
Hazard ratio for the primary end point in SHIFT, by change in heart rate by day 28 (shown by Böhm at Heart Failure Congress 2011)
| Day 28 |
HR (95% CI) |
p |
| Before adjustment for heart rate change |
0.82 (0.75-0.87) |
0.0001 |
| After adjustment for heart rate change |
0.95 (0.85-1.06) |
0.352 |
His arguments, he said, indicate that heart-rate reduction beyond what is achieved with beta blockers can further improve heart-failure outcomes, "that there is a component of heart rate that is apparently independent of beta-blocker treatment, that the only determinant of poor prognosis on beta blockers is still heart rate, and therefore it makes sense to further reduce [heart rate]."
| Ponikowski has previously reported consulting for Amgen and consulting for and receiving honoraria for speaking from Vifor Pharma. To heartwire he said he was scheduled at the current HFA sessions to speak at a Servier-sponsored satellite symposium. Böhm has previously disclosed receiving grants for clinical research and educational activities and consulting for Sanofi-Aventis and receiving research support from Servier. Massie had no disclosures. |
Sources
1.Swedberg K, Komajda M, Böhm M, et al. Ivabradine and outcomes in chronic heart failure (SHIFT): A randomised placebo-controlled study. Lancet 2010; 376:875-885.
2.McAlister FA, Wiebe N, Ezekowitz JA, et al. Meta-analysis: beta-blocker dose, heart rate reduction, and death in patients with heart failure. Ann Intern Med 2009; 150:784-794. |