Beta-Blocker Continuation in Decompensating Heart Failure: A Slow Pulse in a Fast Fever
A single patient, two days into a pneumonia-triggered decompensation of her chronic heart failure, whose fever should be raising her heart rate and isn't — because the same beta-blocker she has taken since her heart attack may also be blunting the compensatory response that would otherwise reveal whether her kidneys and hands are underperfused.
J.M., a 71-year-old woman who has taught a Tuesday-evening watercolor class at her senior center for nearly a decade, missed this week's session — the first time in six years she has cancelled at all. She has ischemic cardiomyopathy from an anterior myocardial infarction four years ago, an ejection fraction that has held around 28% on quadruple guideline-directed therapy, and chronic kidney disease stage 3b with a baseline creatinine near 1.3. Four days ago she developed a fever and a productive cough; yesterday, too breathless to climb her own porch steps, she was admitted with community-acquired pneumonia superimposed on acutely decompensated heart failure, and started on intravenous antibiotics alongside diuresis.
She is now on her second hospital day, and her vitals are not behaving the way this picture should predict. Her temperature is still 100.9°F, but her heart rate is 58 — the reflex tachycardia a fever this real should provoke simply isn't there. Her hands and feet are cool, her systolic pressure has drifted from an outpatient baseline near 112 down to 92, and her creatinine, 1.4 on admission, has climbed to 1.6 in the day since. None of this proves she is underperfused rather than simply dehydrated and unwell — but her carvedilol, 25 mg twice daily since her infarction, blunts exactly the compensatory tachycardia that would otherwise announce low cardiac output more plainly, and its added alpha-1 blockade contributes vasodilation that a purely beta-1-selective agent would not. The question the team is working through this morning is whether that same drug — four years of protecting a heart that has already failed once — is now the reason the earliest signs of trouble are this hard to read, and whether continuing it, cutting it in half, or stopping it outright changes which way the next day goes.
At the bedside, hospital day two
Reduce the carvedilol to twelve and a half milligrams twice daily — don't stop it, and don't switch it. She's four years out from an anterior infarct with reduced ejection fraction, exactly the population the pivotal post-MI trial for this drug enrolled when it showed a mortality benefit, and the discontinuation registry data cut the same direction: patients who came off their beta-blocker during a heart failure admission did worse after discharge than patients who stayed on some dose, even accounting for how sick they looked at the time. Cutting the dose in half buys most of the hypotension relief without asking her heart to relearn how to handle unopposed catecholamines.
I'd stop it, at least for the next day. The registry signal you're citing is largely patients who looked sick, not necessarily patients who looked underperfused the way she does right now — cool extremities, a creatinine that's climbing, a pressure that's fallen twenty points from her own baseline. The trial that actually randomized continuation versus withdrawal in decompensated heart failure excluded exactly this picture; it never tested what to do with a patient who has real signs of hypoperfusion, only patients who were volume-overloaded but still adequately perfused. I don't think a half-dose threads that needle — either the negative inotropy is a problem right now or it isn't, and if it is, twelve and a half milligrams is still not zero.
There's a version of this that doesn't require deciding whether she's truly underperfused today. Carvedilol isn't a clean beta-1 story — its alpha-1 blockade is doing real vasodilating work of its own, on top of whatever the beta-1 component is doing to her stroke volume, and that alpha-1 effect is exactly the part with the weaker claim to a protective role in her. Halving the dose lowers both together, which is why I don't think it fully resolves the disagreement either. If her pressure and her creatinine haven't turned around by tomorrow, the more targeted move is switching her outright to a beta-1-selective agent at an equivalent dose — she keeps the receptor-blockade benefit the mortality trial was actually built on, and loses the alpha-1 contribution that may be the real driver of what we're seeing. I'd hold that in reserve rather than do it today; a same-drug dose reduction is the smaller, more reversible move to try first.
Agreed for the next twenty-four hours: carvedilol reduced to 12.5 mg twice daily, not discontinued; diuresis and antibiotics continue unchanged; blood pressure, creatinine, and perfusion reassessed at the next dosing interval.
Not agreed, and the reason tomorrow's note will need to record which fork this actually turned out to be:
Uptitrate back toward her home dose of 25 mg twice daily within the week, once two consecutive reassessments confirm she's genuinely turned the corner.
Hold the beta-blocker entirely and pursue further evaluation for a true low-output state, with dobutamine available if the picture worsens rather than merely persists.
The clinical pharmacologist's alternative — that persistent hypotension without improvement might argue for switching to a beta-1-selective agent rather than simply resuming or withholding carvedilol — was heard and left on the table for that conversation, not decided today.