Vasospastic Angina With Reduced Ejection Fraction: Two Contraindications, One Patient
Two guideline-level cautions apply to the same patient in opposite directions: the calcium channel blockers best studied for coronary spasm are classified as harmful at her ejection fraction by current heart failure guidance, and the beta-blocker she was about to start carries an older warning against use in spasm. Neither drug list accounts for the other diagnosis.
E.W., a 57-year-old woman, has spent the year since her younger son left for college getting used to a quieter house — most evenings now it's just her and her husband, splitting the newspaper and disagreeing about what to watch. Ten weeks ago that routine stalled when a cardiologist told her the breathlessness she'd been blaming on an empty nest and a slow summer wasn't either of those: an echocardiogram showed a left ventricular EF of 33%, with mild global hypokinesis and no culprit lesion on the coronary angiogram that followed — nonischemic dilated cardiomyopathy, cause still unclear. She quit smoking eight years ago after a fifteen-pack-year history, has well-controlled hypothyroidism on a stable levothyroxine dose, and had no cardiac history before this diagnosis. Her cardiology team started sacubitril/valsartan, spironolactone, and empagliflozin over the following month, titrating one at a time and confirming she stayed euvolemic at each step; a beta-blocker was the deliberately deferred fourth pillar of GDMT, planned for this visit once the first three were settled.
That plan changed three weeks ago. She began waking between two and five in the morning with a tight, central chest pressure that built over a minute or two and eased on its own within fifteen, until a third episode sent her to the emergency department, where a troponin drawn during the pain came back undetectable and an ambulatory monitor placed afterward caught a fourth episode: two millimeters of transient ST-elevation in the inferior leads that normalized within four minutes of a single sublingual nitroglycerin tablet, in step with her pain resolving. A repeat coronary angiogram again showed nothing to explain it directly — no stenosis greater than 20%, no plaque rupture — which is itself part of what points toward vasospasm rather than atherosclerotic disease: transient, exertion-independent ST-elevation with a clean angiogram and a prompt nitrate response is close to the textbook description of vasospastic (Prinzmetal) angina, and her nicotine history is a recognized predisposing factor even eight years out from her last cigarette. The timing collided badly with her own treatment plan. The calcium channel blocker best studied for suppressing coronary spasm, diltiazem, belongs to the non-dihydropyridine group current heart failure guidance classifies as harmful at a reduced EF, for its negative inotropic effect. And the beta-blocker she was about to start carries its own, older caution in vasospastic disease: unopposed alpha-mediated coronary vasoconstriction once beta-2-mediated vasodilation is blocked alongside it. She now has two guideline-level cautions pointing in opposite directions, and no drug on either list was chosen with the other diagnosis in mind.
Ten weeks post-diagnosis, three weeks into spasm
Start amlodipine today and hold the beta-blocker. Diltiazem has the better track record against coronary spasm of the calcium channel blockers, but it's a non-dihydropyridine, and that's exactly the group current heart failure guidance classifies as harmful at her ejection fraction — so amlodipine, not diltiazem, is the CCB we can actually use. That's the easy half of this. The beta-blocker is where I'd rather not move today: she's three of four pillars in already, tolerating all of them, and nothing about her presentation is unstable enough that the fourth pillar has to start on this exact visit.
I'm not arguing beta-blockers are permanently off the table for her — only that starting one in the same visit we're diagnosing active coronary spasm is a worse time to find out whether it provokes another episode than a visit two weeks from now would be.
The amlodipine choice I agree with completely. The beta-blocker deferral is where I'd push back. Every week without one is a week of unrealized mortality benefit in a patient who's otherwise ready for it, and the mechanism behind the deferral — unopposed alpha-mediated vasoconstriction once beta-2-mediated dilation is blocked — is a claim about specific receptor pharmacology, not a blanket property of every drug in the class. Carvedilol blocks alpha-1 receptors directly, at a clinically meaningful magnitude, not as a footnote. That's the specific mechanism the deferral argument depends on, addressed by the specific drug choice.
I'd rather start carvedilol cautiously today than wait on a caution that, on the actual pharmacology of this agent, may not apply to it the way it applies to a plain beta-1-selective drug.
The alpha-1 argument for carvedilol is real pharmacology, not a rationalization — it's a racemic drug with alpha-1 blockade at a magnitude that's been directly measured, not assumed. But it's receptor pharmacology applied to this scenario, not an outcome anyone has actually tested in vasospastic disease specifically, and I'd rather not treat a mechanistically sound argument as equivalent to a proven one. I don't think the two positions here are actually as far apart as they sound: start amlodipine today for the spasm itself, and use the next two weeks as the window to see whether it's controlled before adding carvedilol — not any beta-blocker from the mortality-benefit class interchangeably, specifically carvedilol, since the alpha-1 mechanism is the entire reason it's a reasonable choice here and a plain beta-1-selective agent like bisoprolol would reintroduce the exact concern we're trying to design around.
Agreed today: start amlodipine 5mg daily now, continue sublingual nitroglycerin as needed, and hold carvedilol for roughly two weeks pending a spasm-frequency reassessment — rather than starting any beta-blocker today, and rather than substituting bisoprolol or another beta-1-selective agent for carvedilol if a beta-blocker is added sooner than planned.
Not agreed, and the reason the reassessment carries a branch point rather than a fixed date:
Carvedilol initiation is deferred further, and amlodipine is uptitrated first before the beta-blocker question is revisited.
Carvedilol is started at 3.125mg twice daily as planned, with the first two follow-up visits scheduled specifically around the beta-blocker titration rather than routine spacing.
Genuinely unresolved beyond that branch: how aggressively to uptitrate carvedilol toward its full mortality-benefit target dose if the starting dose proves spasm-neutral. Higher doses carry more of both the intended benefit and the theoretical risk at once, and no one at this visit was willing to commit to a titration ceiling in advance — that decision was left to be made visit by visit, against how she actually responds, rather than fixed today.