Efficacy and safety of renal denervation in addition to pulmonary vein isolation for atrial fibrillation and hypertension—Systematic review and meta‐analysis of randomized controlled trials

Abstract Introduction This systematic review and meta‐analysis aimed to assess the latest evidence on the use of renal denervation (RDN) + pulmonary vein isolation (PVI) compared to PVI alone for treating atrial fibrillation (AF) with hypertension. Methods A systematic literature search from several electronic databases was performed up until January 2020. The primary outcome was AF recurrence defined as AF/atrial flutter (AFL)/atrial tachycardia (AT) ≥30 seconds at 12‐month follow‐up and the secondary outcome was procedure‐related complications. Results There were 568 subjects from five studies. AF recurrence was 90/280 (32.1%) in the RDN + PVI group and 142/274 (51.8%) in the PVI group. RDN + PVI was associated with a lower incidence of AF recurrence (RR 0.62 [0.51, 076], P < .001; I 2: 0%). Pooled analysis of HR showed that RDN + PVI was associated with reduced AF recurrence (HR 0.51 [0.38, 0.70], P < .001; I 2: 0%). Complications were 7/241 (2.9%) in the RDN + PVI group and 8/237 (3.4%) in the PVI group. The rate of complications between the groups was similar (RR 0.87 [0.33, 2.29], P = .77; I 2: 0%). In the subgroup analysis of paroxysmal AF, RDN + PVI was shown to reduce AF recurrence (RR 0.64 [0.49, 0.82], P < .001; I 2: 0% and HR 0.56 [0.38, 0.82], P = .003; I 2: 0%) compared to PVI alone. RDN + PVI has a moderate certainty of evidence in the reducing AF recurrence with an absolute reduction of 197 fewer per 1000 (from 254 fewer to 124 fewer). Conclusion RDN in addition to PVI, is associated with reduced 12‐month AF recurrence and similar procedure‐related complications compared to PVI alone.


| INTRODUC TI ON
Atrial fibrillation (AF) and hypertension are highly prevalent diseases that frequently coexist. 1,2 Hypertension increases the risk of AF, 3 and uncontrolled hypertension has been shown to be an independent predictor of AF recurrence postablation. 4 Despite improvement in technologies, the rate of AF recurrence postablation remained high, [5][6][7] and uncontrolled hypertension poses as an additional problem.
Overactivation of the renin-angiotensin-aldosterone system (RAAS) leads to a heightened sympathetic tone, which in turn stimulates renin synthesis. 8 An enhanced RAAS leads to remodeling along myocardium and vasculature, promoting arterial hypertension and left atrial remodeling. 8 Renal denervation (RDN) may attenuate the hyperactive sympathetic nervous system which in turn leads to a reduction in RAAS activation and lowering of blood pressure. 9,10 RDN with or without PVI has been shown to reduce AF burden in hypertensive patients. 11,12 Hence, RDN as an additional intervention to PVI is potentially advantageous to lower blood pressure (especially in uncontrolled hypertension) and reduce the incidence of AF recurrence. This systematic review and meta-analysis aimed to assess the latest evidence on the use of RDN + pulmonary vein isolation (PVI) compared to PVI alone for treating atrial fibrillation with hypertension.

| Search strategy
We performed a systematic literature search on topics that compare the use of RDN in addition to PVI and PVI alone for hypertensive AF patients undergoing catheter ablation with keywords ["renal denervation" and "ablation" and "atrial fibrillation"] and its synonym from inception up until January 2020 through PubMed, EuropePMC, Cochrane Central Database, ScienceDirect, ProQuest, ClinicalTrials. gov, and hand-sampling from potential articles cited by other studies. The records were then systematically evaluated using inclusion and exclusion criteria. We also perform hand-sampling from references of the included studies. Two researchers independently performed the initial search and discrepancies were resolved by dis-

| Selection criteria
The inclusion criteria for this study are studies that compare the use of RDN in addition to PVI and PVI alone for hypertensive AF patients undergoing catheter ablation. We include RCTs and exclude observational studies, animal studies, case reports, review articles, and non-English language articles.

| Data extraction
Data extraction and quality assessment were done by two independent authors using a standardized extraction form that includes the first author, the year of publication, study design, sample size, AF recurrence/freedom, complications, paroxysmal AF, age, gender, baseline systolic blood pressure, and follow-up duration.
The primary outcome was AF recurrence defined as the presence of AF/AFL/AT ≥30 seconds at 12-month follow-up and the secondary outcome was procedure-related complications.

| Statistical analysis
To perform the meta-analysis, we used RevMan version 5.3 software (Cochrane Collaboration). Risk ratios (RRs) were used for pooled effect estimate of dichotomous data, the hazard ratio (HR) was used for pooled effect estimate for HR, and their 95% confidence interval. Inconsistency index (I 2 ) test, which ranges from 0% to 100%, was used to assess heterogeneity across studies. A value above 50% or P < .10 indicates statistically significant heterogeneity. We used the Mantel-Haenzsel method to calculate RRs and generic inverse variance to calculate pooled HR using a random-effects model for meta-analysis regardless of heterogeneity. Subgroup analysis was performed to evaluate outcome in patients with paroxysmal AF. All P values were two-tailed with a statistical significance set at 0.05 or below. Cochrane risk-of-bias tool for randomized trials (Cochrane Collaboration) will be used to assess the risk of bias.
The certainty of evidence was assessed using Guideline Development Tool by GRADEpro GDT (McMaster University and Evidence Prime Inc.).

| Risk of bias assessment
The risk of bias was mainly due to performance bias because the operators were not blinded to the groups, due to the fact that the same operator performed both PVI and RDN. ( Figure 4A) The risk for selection bias was unclear, the method of randomization and allocation concealment was not clearly defined in some studies. The funnel-plot was symmetrical for the AF recurrence outcome indicating a low-risk of publication bias. ( Figure 4B).

| GRADE approach
Grading of Recommendations Assessment, Development, and Evaluation (GRADE) showed a moderate certainty of evidence that RDN in addition to PVI results in the reduction of AF recurrence with an absolute reduction of 197 fewer per 1000 (from 254 fewer to 124 fewer). (Table 2). In animal models, RDN has been shown to reduce catecholamine levels, reverse atrial structural, and electrical remodeling. 22,23 In animal ischemic heart failure models, RDN has been shown to reverse structural and electrical remodeling of the atrium along with reduced AF inducibility. 24 RDN has been though to reduce sympathetic overactivity and arrhythmogenic foci/substrate arising from it. 12 RDN alone without PVI has been shown to reduce AF burden and increase the quality of life in patients with AF. 11 Furthermore, RDN in PVI patients was shown to reduce left ventricular septal thickness and posterior wall thickness. 13,16 Abnormal interventricular septal thickness has been shown to increase the risk of mortality and stroke in AF patients, and the regression of left ventricular hypertrophy has been hypothesized as a therapeutic target. 25 While shorter diagnosis-to-ablation time has been shown to reduce AF recurrence in PVI, 26 it is not known whether time to RDN will influence its potential to reverse the pathology.

| D ISCUSS I ON
This meta-analysis showed that RDN + PVI was associated with reduced AF recurrence. Romanov et al reported that RDN in addition to PVI was independently associated with less AF recurrence. 15 There seemed to be no statistically significant difference between those receiving high-frequency stimulation compared to those without. 16 Besides the direct attenuation of sympathetic effect on the atrium, RDN has also been shown to reduce AF recurrence through the reduction of BP in patients with uncontrolled/drug-resistant hypertension. Romanov et al study showed that the AF burden in RDN + PVI was lower compared to PVI only (2.43% vs 6.95%). 15 Their study also showed that the blood pressure reduction was associated with less AF burden as demonstrated by the increase in a mean difference of AF burden between the two groups later in follow-up period. Furthermore, AF recurrence was also shown to be decreased by a HR of 0.9 with each 5 mm Hg decrease in blood pressure. 15 Kiuchi et al study demonstrated an inverse correlation between blood pressure reduction and AF burden. 12 This is unsurprising because uncontrolled hypertension (despite ≥3 antihypertensive drugs) has been shown to independently increase the risk of AF recurrence post-ablation, while controlled hypertension did not affect ablation outcome. 4 The aforementioned study found that increased atrial size, nonpulmonary vein triggers, and extensive atrial scars were more frequently found in uncontrolled hypertensive group compared to controlled hypertensive, explaining the higher AF recurrence. 4 Steinberg et al showed that RDN + PVI was associated with reduction in left atrial size which may contribute to a lower AF recurrence. 16

| Limitations
The limitation of this systematic review and meta-analysis is mainly due to publication bias in which positive studies are more likely to be published than negative studies. Although the funnel-plot analysis did