The Evaluate Renal Artery Denervation in Addition to Catheter Ablation to Eliminate Atrial Fibrillation (ERADICATE-AF) trial: arrhythmia suppression, blood pressure control, and potential interactions

Graphical Abstract

Graphical Abstract.

The Evaluate Renal Artery Denervation in Addition to Catheter Ablation to Eliminate Atrial Fibrillation (ERADICATE-AF) trial was the largest randomized clinical trial that tested the clinical impact of renal artery denervation (RDN) in conjunction with pulmonary vein isolation (PVI) for atrial fibrillation (AF) compared with PVI alone.¹ PVI + RDN demonstrated significantly greater freedom from 12-month AF recurrence > 30 s in comparison with PVI alone.¹ The results have been affirmed in several recently published meta-analyses.^2–3 In this report, we present the analyses of the prespecified and additional secondary endpoints of the ERADICATE-AF trial that included details of blood pressure (BP) control and relationship to the antiarrhythmic effect of RDN, AF burden as an arrhythmia endpoint, and subgroup effects.

The ERADICATE-AF trial was a single-blind, randomized clinical trial that enrolled 302 patients with paroxysmal AF who had suboptimally controlled hypertension. At the follow-up 7-day Holter at 12 months, the AF burden was significantly lower in the PVI + RDN group: 1.43 ± 3.36% (median, 0; range, 0 to 21), in the PVI + RDN group vs. 3.90 ± 8.53% (median, 0; range, 0 to 60), and in the PVI only group (P = 0.0012). If AF recurrence was defined as arrhythmia events >4 h (which is much more correlated with future AF burden than 30 s cut-off),⁴ freedom from AF was 69.2% in the PVI only group vs. 84.7% in the PVI + RDN group at 12 months (P = 0.002). Risk of AF lasting for longer than 4 h was significantly reduced by RDN: hazard ratio (HR) 0.45 [95% confidence interval (CI), 0.27–0.76, P = 0.003]. None of the patients developed persistent AF (i.e. duration of AF >7 days). Development of first AF was associated with an average 6.2% increase in AF burden (95% CI, 4.9–7.4%, P < 0.001).

The degree of improvement in systolic blood pressure (SBP) from baseline to 12 months in the PVI + RDN group was used to create four subgroups, and the AF burden at 12 months was compared among them. Relative to ΔSBP in the RDN + PVI group, all subgroups had similar low AF burdens (P = 0.11) (Graphical Abstract).

Treatment effect was consistent across all measured subgroups, based on non-significant interaction P-values. For example, the patients’ background medical history or treatment with beta blockers, or the patients’ baseline degree of hypertension control, did not affect the favourable response to RDN.

Univariate and multivariate predictors of the primary endpoint are presented in Table 1. Only the performance of RDN (multivariate HR 0.55) was independently associated with a reduction in the likelihood of AF recurrence, while a history of congestive heart failure was associated with increased risk (multivariate HR 1.63).

Table 1.

Univariate and multivariate predictors of primary endpoint

	Univariable analysis		Multivariable analysis
	HR (95% CI)	P-value	HR (95% CI)	P-value
Renal denervation	0.57 (0.38–0.85)	0.003	0.55 (0.37–0.81)	0.002
Congestive heart failure history	1.61 (1.01–2.60)	0.049	1.63 (1.01–2.65)	0.048
Interventricular septum thickness (+1 mm)	0.88 (0.78–0.99)	0.042	0.89 (0.78–1.01)	0.06
Age (+1 year)	0.99 (0.97–1.02)	0.68
Male sex	1.11 (0.75–1.64)	0.60
History of stroke/TIA	0.93 (0.48–1.78)	0.81
History of diabetes	1.42 (0.82–2.45)	0.21
CHA2DS2-VASc (+1 score)	1.00 (0.87–1.15)	0.98
Duration of atrial fibrillation (+1 year)	0.90 (0.74–1.10)	0.30
Baseline systolic blood pressure (+5 mm Hg)	0.94 (0.84–1.04)	0.24
Baseline diastolic blood pressure (+5 mm Hg)	0.96 (0.84–1.10)	0.58
Baseline creatinine (+5 µmol/L)	1.06 (0.97–1.16)	0.21
Atrial arrhythmia on baseline ECG	0.41 (0.02–7.40)	0.54
Antiarrhythmic drug therapy pre-ablation	0.90 (0.50–1.61)	0.72
Left ventricular ejection fraction (+5%)	0.93 (0.78–1.12)	0.45
Left atrium size (+5 mm)	0.96 (0.71–1.30)	0.78
Atrial flutter ablation	0.69 (0.38–1.23)	0.20

CI, confidence interval; ECG, electrocardiogram; HR, hazard ratio; TIA, transient ischaemic attack.

Among patients with symptomatic paroxysmal AF with a background history of suboptimally controlled hypertension who were referred for catheter ablation, this trial demonstrated that PVI + RDN significantly increased the likelihood of freedom from AF over 12 months compared with PVI alone, whether using a conventional definition of AF >30 s or the more powerful and clinically relevant endpoint of AF >4 h.⁴ Notably, favourable response to RDN was independent of effects on BP control, suggesting a potent impact by another mechanism. The use of RDN in ERADICATE also demonstrated significant secondary benefits, including reduced AF burden overall, enhanced BP control,¹ and favourable echocardiographic atrial and ventricular remodeling.¹ Renal artery denervation was accomplished with an excellent short- and long-term safety record and was an independent powerful predictor of freedom from AF in comparison with PVI alone. The benefits of RDN on AF control were consistent across all measured subgroups.

Improved AF outcome after RDN may have been mechanistically due to (i) better BP control,⁶ (ii) a direct antiarrhythmic effect mediated by sympatholysis, or (iii) both. However, RDN's antiarrhythmic effects are seen across the spectrum of resistant to milder hypertension, in the setting of ventricular arrhythmias⁵ and other conditions such as sleep apnoea⁷ and heart failure.⁸ This suggests that a unifying mechanism of therapeutic action could be the reduction of central sympathetic output. Consistent with this proposed mechanism, we found that the improved post-ablation outcome was not dependent on improved BP control, suggesting a direct positive effect of RDN on post-ablation AF recurrence. Further, all subgroups responded to RDN, irrespective of baseline BP control or use of beta-adrenergic blockers.

In light of these findings and implications, a new trial, ERADICATE-AF II, is now being launched to test RDN as adjunctive procedure during ablation for persistent AF and either controlled hypertension or no hypertension (NCT0511638).

In this multicentre randomized trial, patients in whom RDN was added to PVI had a significantly lower likelihood for AF recurrence using different definitions of AF duration and as well as AF burden at 12 months than patients with PVI alone. Atrial fibrillation burden was similarly reduced independent of improvement in SBP after RDN suggesting that BP reduction is not responsible for the antiarrhythmic response.

Data Availability

The data underlying this article will be shared on reasonable request to the corresponding author.