Ivabradine in HFrEF: Rate Control at the Beta-Blocker's Ceiling
His beta-blocker is not undertreated — it's genuinely maxed out, confirmed twice over. His resting heart rate still won't come down. The disagreement isn't about whether to add ivabradine; it's about how much of its trial evidence actually belongs to a patient like him.
J.C., a 58-year-old man who has spent three decades running the CNC lathes at a small precision machine shop, first noticed something was wrong the way tradesmen usually do — not from a single symptom, but from a habit quietly breaking. He had always worked a full shift standing at the same machine; sometime last year he started finding reasons to sit down partway through it, telling himself it was the heat in the shop rather than his own chest. Hypertension had been on his chart since his mid-forties, well controlled for over a decade on a single agent and never the reason he saw a doctor more than once a year. The dyspnea and ankle swelling that eventually sent him to a cardiologist fourteen months ago were new, and the workup that followed was unhurried but thorough: clean coronaries on catheterization, no meaningful alcohol history, nothing on further testing to explain the finding beyond nonischemic dilated cardiomyopathy, ejection fraction 25% at diagnosis.
Fourteen months of guideline-directed therapy have done real, measurable work. His ejection fraction has climbed to 32%, his NT-proBNP has fallen by roughly two-thirds, and he is back to working a full shift most days. What hasn't moved, across six months of serial clinic visits, is his resting heart rate — steady in the high seventies to high eighties despite a regimen that already includes carvedilol, sacubitril/valsartan, spironolactone, and dapagliflozin. Twice, three months apart, his cardiologist tried to push the carvedilol past 12.5 mg twice daily, and twice he came back reporting the same thing: standing up too fast at the lathe left him gray and reaching for the workbench. Twelve-point-five milligrams twice daily is not an arbitrary stopping point for him; it's the dose his own physiology has set, confirmed twice rather than assumed once. An elevated resting heart rate in a ventricle already working at reduced capacity is not a cosmetic finding — it shortens diastolic filling time and raises myocardial oxygen demand on a heart with little margin to spare — and the question in front of today's visit isn't whether that number matters, but what's left to do about it once the drug usually reached for first has already reached the ceiling this patient's own body set for it.
At the follow-up visit, six months into a plateaued heart rate
Start ivabradine today. He meets every element of the indication as written: ejection fraction 32%, comfortably under the 35% threshold; sinus rhythm confirmed on Holter and twelve-lead; resting heart rate above 70 on three consecutive visits, not a single reading; and, critically, he's on a maximally tolerated beta-blocker dose, not an undertreated one — that's a real, twice-demonstrated ceiling, not a default nobody pushed against. SHIFT showed an 18% relative reduction in the composite of cardiovascular death or heart-failure hospitalization, driven mainly by fewer hospitalizations, and the mechanism tracks with what we're already seeing on his own echo — heart-rate reduction with ivabradine reverses left ventricular remodeling independent of beta-blocker dose or the original cause of the cardiomyopathy. Every other number is already trending the right direction. This is the one lever still on the table.
I don't disagree with starting it — I want to be precise about why, because the trial's own numbers are more complicated on this exact point than the topline result suggests. Only 49% of SHIFT's roughly 6,500 patients were even at half of their target beta-blocker dose at baseline, despite repeated efforts from the trial's own steering committee to push doses higher. A meaningful share of that 18% relative-risk reduction almost certainly reflects heart-rate lowering in patients who were undertreated on beta-blockade to begin with, not patients like him who are genuinely at a documented physiologic ceiling. I'm not confident the same effect size applies to a population that's already maximized rather than under-dosed — that's a real, unresolved gap in the evidence, not a reason to withhold the drug, since nothing in the trial's eligibility criteria excluded patients like him either.
If he were a patient who'd simply never had his carvedilol pushed past a starting dose, I'd want that tried honestly before reaching for a second agent. That's not this case — two independently reproduced episodes of orthostatic symptoms at the next dose step is a real ceiling, not a soft one.
Mechanically, there's nothing in the way. He isn't on a strong CYP3A4 inhibitor, so no dose adjustment is needed there, and digoxin was never a serious alternative here — in true sinus rhythm its chronotropic effect is weak and unreliable, nothing like the AV-nodal rate control it provides in atrial fibrillation. Ivabradine itself measurably raises the risk of new atrial fibrillation — 5.0% per patient-year against 3.9% on placebo in SHIFT — which is a real irony given the drug only works in sinus rhythm to begin with. That belongs in the monitoring plan, not as an afterthought. He should also be told about phosphenes before he leaves today — reported by about 3% of patients on ivabradine, usually a brief brightness at the edge of the visual field, rarely treatment-limiting. If he doesn't know to expect it, he's more likely to call it in as something alarming than shrug it off.
Before anyone titrates anything, I want the mechanism confirmed, not assumed. A resting tachycardia in the high seventies to high eighties in a structurally abnormal heart isn't automatically sinus in origin — it can be an ectopic atrial focus running at a similar rate, and ivabradine's target is the sinoatrial node's own If current specifically. If this isn't genuine sinus-node automaticity, the drug simply won't do what we're asking it to do. His Holter and resting twelve-lead both show normal P-wave axis and morphology consistent with sinus origin, so I'm satisfied the mechanism is right here — but that check has to happen before the prescription, not be assumed from the number alone.
On the beta-blocker question, I'd start ivabradine today rather than ask for a third attempt at a dose he's already failed twice — two independently reproduced episodes is a materially different patient than the SHIFT enrollee who simply never had a serious attempt made. Standard dosing applies; he's not in the reduced-starting-dose group for age or baseline bradycardia, so 5 mg twice daily, reassess heart rate at two weeks, target range 50 to 60. Given the atrial fibrillation numbers, that two-week visit needs a rhythm strip, not just a heart rate, and every visit after it for the first several months. What I won't pretend to resolve is how much of SHIFT's 18% actually belongs to a patient at a true ceiling rather than an undertreated one — that's a genuinely open evidentiary question, and today's decision doesn't answer it.
Agreed within the visit: ivabradine started today at the standard 5 mg twice daily, carvedilol continued unchanged at 12.5 mg twice daily rather than attempted a third time, and a two-week follow-up scheduled with both a resting heart rate check and a rhythm strip — not a heart rate alone — given the drug's own measured effect on atrial fibrillation incidence.
Not agreed, and left explicitly open rather than resolved by consensus:
The 18% relative risk reduction applies to him as directly as it did to the trial population, and today's addition is exactly the benefit the evidence supports.
Some real fraction of that number belonged to patients who were undertreated on beta-blockade, not maximized like he is, and the true effect size for a patient at a genuine ceiling remains genuinely unknown.
Nobody at the table treated this as a reason to wait — ivabradine was started today regardless — but the disagreement over how much of SHIFT's numbers to expect for this specific patient stays on his chart as an open question, to be revisited once his own two-week and subsequent heart-rate and rhythm data are in hand.