CABANA (Catheter ABlation vs. ANtiArrhythmic drug Rx
for AF)
HRS 05.2018

Study designed to compare catheter ablation vs medical Rx in all patients.

More than 2200 randomized patients showed that catheter ablation was no worse than drugs for
combined endpoint, or for the solitary endpoint of all cause mortality.
15% were > 75 years old.
23% with OSA
15% with heart failure
10% with prior CVA or TIA
43% with parxysmal AF and 47% with persistent AF (10% with long term persistent AF)

Failed to accomplish primary endpoint: not superior to medical management for a combined endpoint
of all-cause death, stroke, serious bleeding, or cardiac arrest.

It did prove that ablation was superior to medical therapy as a more definitive treatment of
atrial fibrillation by cutting the rate of recurrent arrhythmia nearly in half.

May have been superior to medical Rx if only randomization assignments had been more closely
followed as the trial proceeded. But that didn't happen, with about
30% of patients assigned to medical
management crossing over to undergo catheter ablation, presumably because they had received
inadequate symptom relief from their drug regimens.
In addition 10% of patients assigned to catheter ablation did not undergo it.
Intention to treat analysis are based on a foundation where most patients are maintained on their
assigned treatment.
These crossovers produced a disparity in the outcomes between the standard intention-to-treat
analysis, which showed a neutral difference between the two study arms.

The per-protocol analysis showed a statistically significant
27% relative risk reduction in the primary
endpoint among the patients randomized to and actually treated with catheter ablation, compared with
those randomized to and exclusively treated medically.

The prespecified on-treatment analysis which instead of censoring crossover patients analyzed
outcomes based on the treatments that patients actually received showed a statistically significant one
third reduction in primary endpoint among the ablated patients and a statistically significant 40%
relative reduction in all-cause mortality in the ablated arm, compared to those treated with medical

If you take all patients including crossovers ablation was linked with a 14% relative reduction in the
composite primary endpoint which was not statistically significant.
All cause mortality was 15% lower in the ablation arm (also not statistically significant)
All cause mortality and cardiac hospitalization was 17% lower in the ablation arm which was statistically
significant (p=0.002)

For symptomatic treatment and to restore sinus rhythm there is no question that ablation is better. We
knew that before the trial and we know it more convincingly now.

There was a better reduction in atrial fibrillation burden with ablation which was quite impressive with
this study.
The quality of life data collected in 2018 will come out in a separate report later this year.
The study showed also that there was no harm with ablation.

The risk of ablation was very low:
Pericardial effusions 2.2% that required no intervention
1.8% pericarditis successfully treated medically with NSAIDs

AF recurrence after 90 days was 47% lower in the ablation arm vs medical Rx

An intriguing subgroup analysis for the primary endpoint was patients younger than 65 years of age
esp those with a history of heart failure, while those who were older than 75 did worse

I recommend that people interested in CABANA look at the full data set and not limit themselves to a
knee-jerk reaction with the intention-to-treat analysis

NEJM 2018 Feb 1;378 [5]:417-27

Study the effectiveness of catheter ablation in patients with heart failure EF <35% (who have ICDs
Primary endpoints: mortality and progression of heart failure
Ablation group 30% paroxysmal 41 % persistent

~40% RRR in primary endpoint in ablation group