Against the Instinct to Act
For the study of the week, cardiologist Mohammed Ruzieh writes about one of my all-time favorite trials--RACE 7 ACWAS.
I can’t say enough good things about Dr. Ruzieh. He trained under Andrew Foy and it shows. He is a keen appraiser of medical evidence. I am delighted that he covered such an important trial. JMM
There is a reflex in emergency medicine and cardiology that goes something like this: patient arrives in atrial fibrillation, the monitor confirms it, and a cardioversion, an electric shock applied externally to the heart, makes it disappear. Problem solved. Everyone goes home happy.
The reflex is intuitive. It is also, it turns out, frequently unnecessary.
At Cardiology Trials Substack, we recently reviewed RACE 7 ACWAS - an important trial that has not received the attention it deserves.
RACE 7 ACWAS, published in the New England Journal of Medicine in 2019, randomized 437 patients presenting to Dutch hospitals with stable, symptomatic, recent-onset AF of less than 36 hours duration to either immediate cardioversion or a wait-and-see approach - rate control, discharge, and cardioversion only if the arrhythmia had not resolved within 48 hours.
The primary endpoint was simple: presence of sinus rhythm at four weeks.
The result: 91% of the delayed group were in sinus rhythm at four weeks versus 94% in the early cardioversion group. A difference of −2.9 percentage points that was within the non-inferiority margin the authors specified. Restoring sinus rhythm quickly does not buy durable rhythm stability. It changes the route, not the destination.
But the primary endpoint is not the main story, in the wait-and-see group, 69% of patients converted spontaneously within 48 hours on rate-control medication alone. Only 28% required a cardioversion.
Meanwhile, in the early cardioversion group, 16% converted on their own before the cardioversion could be performed. And 20% of otherwise eligible patients never made it into the trial at all because they had already returned to sinus rhythm in the waiting room. Read those numbers again. Most patients with recent-onset AF will convert without any intervention beyond rate control.
Cardioversion is not a minor procedural inconvenience. It requires sedation, monitoring, physician time, and carries a small but real risk of complications. It is not nothing.
What the trial cannot tell us: RACE 7 ACWAS was not powered for hard outcomes such as stroke, death, or hospitalization. The small event numbers in both arms are reassuring but not conclusive. The population is also worth noting. Of 3,706 screened patients, 2,581 (70% ) were ineligible, most often because the AF had been going on longer than 36 hours or there was a history of persistent atrial fibrillation. RACE 7 trial results apply to the stable, uncomplicated, early-presenting patient. Not the unstable patient, not the patient with atrial fibrillation of unknown duration, not the patient in whom rate control fails.
Why this still matters in 2026: The trial was published in 2019. The findings are arguably more relevant now than they were then.
Wearables and digital devices are detecting AF in patients who may be asymptomatic or minimally symptomatic, prompting emergency visits for arrhythmias that might have gone unnoticed, and untreated, without consequence a decade ago. The pressure to do something when a monitor shows AF is considerable. RACE 7 ACWAS is a useful antidote to that pressure.
For many patients with new AF, the wait-and-see approach is not a passive strategy. It is an active decision to avoid a procedure most patients don’t need, backed by a randomized trial.
Similar Story with Premature Ventricular Complexes (PVC)
The same principle extends beyond atrial fibrillation. In patients with frequent premature ventricular contractions, a study found that 44 of 100 patients with a median PVC burden of 18.4% experienced spontaneous reduction to below 1% at a median of 15.4 months - without any suppressive therapy.
And this is not unique to cardiology. Most people do not present to a physician when they develop flu-like symptoms, abdominal pain, or diarrhea, and for many, waiting a few days at home is entirely appropriate.
Good medicine is not defined by how much is done, but by the wisdom to know when action is warranted and when restraint is the better prescription. The two are equally important.
 | A guest post by
|
An excellent piece with a well-taken message.
One tiny quibble with word choice: we often say the patient “required” a treatment or intervention when we mean that the patient “was provided” with a treatment.
The implication is that, had we not provided a treatment, an undesirable outcome would have occurred. In the AF study patients who “required” cardioversion after 48 hrs, we don’t know whether they would have spontaneously converted subsequently. They were provided with cardioversion because that is what the study protocol prescribed. Let’s try to distinguish “Patient needed the treatment” from “Patient was given the treatment”.
As this piece so well describes, often our decision to provide a patient with a treatment arises from factors in us, not the patient.
A testimony to the often ignored power of the human body to “heal thyself “.