Amiodarone vs. Dronedarone: Choosing a Rhythm Drug for Thirty More Years of Ladders
A locksmith in his late sixties, two self-terminating atrial fibrillation episodes in three months, and hypertensive structural change nowhere near what shaped either drug's own worst reputation. The choice turns less on which drug controls rhythm better today than on which one he can safely carry for decades.
R.K., a 68-year-old man, has run his own locksmith business for over thirty years — work that still has him climbing extension ladders and folding himself into tight spaces under dashboards several days a week, and that he has no plan to give up. A routine echocardiogram eight months ago, ordered to work up an innocent murmur, incidentally showed mild concentric LVH and mildly impaired relaxation: unsurprising after fifteen years of hypertension, even hypertension that has stayed reasonably controlled on lisinopril and amlodipine, but a first real look at what that history has done to his heart. Three months ago he felt his heart "take off" while re-keying a lock at the top of a ladder, dizzy enough that he climbed down and sat on the truck's tailgate until it passed; an ED visit that afternoon caught AF at a rate of 132, which converted on its own before anyone treated it. A second, shorter episode two weeks ago did the same. He has no coronary disease, no prior heart failure, and an EF of 60% on that same echocardiogram — real structural change, but nothing close to the systolic or decompensated failure that shaped how the two rhythm-control drugs now in front of him were actually studied.
Amiodarone and dronedarone both suppress AF by prolonging the atrial action potential, but they diverge sharply in what that mechanism costs the rest of the body. Amiodarone's iodine-rich structure and long tissue half-life make it the more effective of the two at keeping a heart in sinus rhythm — the DIONYSOS comparison put real numbers behind that reputation — and are also exactly what drive its thyroid, pulmonary, and hepatic toxicity over years of use. Dronedarone's shorter half-life and non-iodinated structure trade away some of that power for a materially cleaner safety profile, a trade ATHENA validated in a population whose structural heart disease looked a great deal like his: roughly three in ten of that trial's patients already carried an NYHA I–III heart-failure label, and dronedarone still reduced cardiovascular hospitalization in that group. What ATHENA excluded, and what makes R.K. genuinely different from the population where dronedarone later caused real harm in ANDROMEDA, is anyone with NYHA IV symptoms or a recent decompensation — ANDROMEDA's patients all had the latter. He is nowhere near that population. He is also, on today's rhythm strip, back in sinus and asymptomatic — which is exactly the moment the group has to decide how much of the rest of his working life it is willing to hand a demonstrably toxic drug versus a demonstrably weaker one.
Cardiology clinic, rhythm-control referral
I'd start amiodarone. Head to head against dronedarone in DIONYSOS, it kept more patients in sinus rhythm — this isn't a marginal preference, it's the better drug at the one job we're actually asking it to do. He's had a near-syncopal episode at the top of a ladder, alone, and a second breakthrough on rate control alone. The stakes of a third episode aren't abstract for this particular patient.
I'm not dismissing the toxicity concern — I'm saying it's manageable with the monitoring we'd put in place regardless, and a drug that actually works is worth that monitoring burden for a man who climbs ladders for a living.
I'd weigh this differently. ATHENA enrolled a population whose structural heart disease looked like his — roughly a third already had NYHA I–III heart failure — and dronedarone still cut cardiovascular hospitalization in that group. The toxicity that makes amiodarone hard to carry for decades — thyroid, lung, liver — isn't a monitoring inconvenience, it's a cumulative cost that gets worse the longer someone stays on it, and R.K. is 68 with a normal life expectancy ahead of him. He sits well clear of the ANDROMEDA population where dronedarone proved harmful — mostly NYHA II–III, but every one of them recently hospitalized for worsening failure with an ejection fraction at or below 35% — and well clear of the permanent-AF population PALLAS flagged — this is exactly the patient dronedarone's safety case was built around.
DIONYSOS's efficacy gap is real and I'm not minimizing it — AF recurrence at twelve months was 63.5% on dronedarone against 42.0% on amiodarone. But that same trial found a significant 39% relative reduction in non-GI adverse events with dronedarone, and that is the trade I'd take for a man who needs a drug for decades.
Both of you are arguing about which antiarrhythmic — I'd ask whether he's actually exhausted rate control first. Two spontaneously-converting episodes in three months on a beta-blocker he's barely titrated is a real pattern, but it's an early one. That said, I take the occupational framing seriously: he works alone, at height, and a third episode with worse timing is a fall risk, not just a rhythm problem. Given that, I'd rather start rhythm control now than watch and wait — and between the two options, dronedarone's safety case fits a man who otherwise has decades of this drug ahead of him.
Agreed: start dronedarone 400mg twice daily, continue metoprolol succinate, obtain baseline TSH and LFTs (already normal today, but re-checked per protocol before the first dose), and repeat both at three months. R.K. also agreed, independent of which drug was chosen, not to work alone at heights until a repeat ambulatory rhythm check confirms he's stayed in sinus rhythm.
Not agreed, and carried forward explicitly rather than smoothed over:
The Electrophysiologist's position is that this confirms amiodarone should have been first-line, and would move to it directly rather than trying a second agent in between.
The Clinical Pharmacologist's position is validated, and no further discussion is needed — but this won't be known for months, not today.
The Primary Care Physician's original question — whether rhythm control was earned yet at all — was resolved in favor of starting it now, on the occupational-risk argument the other two voices supplied, not on a change in her own read of his rhythm history.