Herein, we present the largest meta-analysis of RCTs exploring the effect of adding SVCI to PVI on AF recurrence in patients undergoing AF ablation. The main findings of the current meta-analysis are as follows:

The addition of empiric SVCI to PVI in patients with PAF is associated with a significant 46% reduction in AF recurrence as compared with PVI alone;

SVCI was not associated with a significant increase in terms of procedural and fluoroscopic times as related to PVI alone;

SVCI did not result in an increased rate of peri-procedural complications compared with PVI alone strategy.

Embryologically, the SVC comes from a communication among the sinus venosus and the right atrium and contains cells with property of automaticity, due to the presence of phase 4 depolarization, and of triggered activity [25]. These cells are in myocardial extensions found inside the SVC, the so-called myocardial sleeves, which are responsible for arrhythmogenesis. In this view, it is not surprising that larger and longer myocardial sleeves are associated with higher arrhythmic properties, as recently confirmed by Dong et al. [24], who found that patients with SVC triggers had significantly longer SVC muscle sleeves as compared with patients without inducible SVC triggers. Of note, SVC is the most common site of origin of non-PV triggers, accounting for 25–40% of all non-PV triggers [24], and plays a role not only as a trigger but also as AF perpetuator. Indeed, in a series of 74 patients with SVC-associated AF, SVC initiated AF in 78.4% of cases, and in 32.4% of patients, SVCI was associated with AF termination, conversion to atrial flutter or persistence of atrial arrhythmias confined to the SVC, suggesting its role as perpetuator [26].

The involvement of SVC in AF strongly depends on AF patterns. As shown by Miyazaki et al. [26], an arrhythmogenic SVC was more prevalent between patients with PAF (8.5%), whereas in persistent and long-standing persistent AF, the prevalence was less than 2%. These findings suggest that SVCI may be especially beneficial in PAF patients and may explain the benefit of SVCI on AF recurrences found by Corrado et al. [21], who randomized 320 patients with paroxysmal, persistent, or permanent AF to receive PVI alone versus SVCI plus PVI. Although there was no difference in AF recurrence rates among the two-treatment groups at 1-year follow-up in the overall population, PAF patients undergone SVCI plus PVI had significantly less AF recurrences than PAF patients randomized to PVI alone [21].

The systematic review and meta-analysis from Sharma et al. [27] comprising the RCTs of Corrado et al. [21], Da Costa et al. [22], and Wang et al. [23] has previously shown a trend toward statistical significance in terms of reduction of AF recurrence solely in the PAF population, while no difference was found in total AF population when comparing SVCI + PVI with PVI alone. Conversely, our meta-analysis including the recent study of Dong et al. [24] showed a significant reduction in AF recurrence risk in PAF population, with a pooled OR of 0.54 ([95%CI 0.32;0.92], p-value 0.02, I2 0%), and a trend toward an AF recurrence reduction in the overall population as well (OR 0.66 [95%CI 0.43;1.00], p-value 0.05, I2 0%). Our meta-analysis reveals for the first time that the addition of empiric SVCI to PVI in patients with PAF is associated with a significant 46% reduction in AF recurrence as compared to PVI alone. The significant reduction in AF recurrences with SVCI found in the pooled analysis totaling 440 PAF patients suggests that the results of previous studies might have been hampered by their small sample sizes. In particular, the study by Wang et al. may have been also influenced by the short follow-up period and by the high incidence of PV reconnections among patients with AF recurrences in the SVCI plus PVI group, yet none of them demonstrated SVCI reconnections [23].

In patients with SVC trigger-induced AF, SVCI has demonstrated an effective ablation strategy. Chang et al. [28] achieved a 73% freedom-from-AF rate at 5-year follow-up after a single SVCI procedure and without PVI in patients with SVC-triggered AF. More recently, Dong et al. [24] found a 93.3% of AF freedom rate at 1-year follow-up in PAF patients with SVC triggers undergone SVCI plus PVI, showing the importance of SVCI in patients with inducible SVC triggers. However, they did not find significant benefit of adding empiric SVCI to PVI in patients without inducible SVC triggers (log-rank p-value 0.28), pointing against an empiric SVCI approach in PAF patients. The last three studies included in our pooled analysis enrolled unselected AF patients, with non-reported or a low rate of provoked SVC triggers elicited with a standard protocol, ranging from 3.1 to 3.7% [21, 23]. As a result, the meta-analysis includes a population without or with a very low rate of SVC-triggers. This consideration emphasizes our report of a significant 46% risk reduction of AF recurrence when empiric SVCI is added to PVI in a population mostly unselected for SVC triggers. Conversely to our results, the study by Dong et al. [24] showed that SVCI was not beneficial in patients without SVC triggers. The study by Dong et al. [24] used an aggressive protocol to elicit SVC triggers, with isoproterenol infusion, rapid burst pacing, and high doses of adenosine, that allows the recognition of higher rates of non-PV triggers (23.1% of patients showed SVC triggers). However, this induction protocol is time consuming and may be difficult to apply in daily clinical practice. Moreover, due to the transient nature of non-PV foci, several non-PV triggers may be unidentified despite aggressive provocative maneuvers, as shown by Miyazaki et al. [26]. In this view, an empiric SVCI approach may be preferred over an as-needed SVCI strategy and may explain our finding of significant AF recurrence reduction with SVCI in patients without or with a low rate of SVC-trigger inducibility.

Importantly, our meta-analysis showed that SVCI is not only effective but also safe. Indeed, no difference in terms of complications rates was found among the two groups (Table A.2), with two phrenic nerve injuries (PNI) occurring in the study by Da Costa et al. [22]. The incidence of PNI is low (0–5%), usually transient, and may be avoided searching phrenic nerve capture before ablation by pacing using high output. Moreover, the right phrenic nerve may be visualized by ICE during AF ablation, thus preventing its injury during radiofrequency delivery [29]. Yamaji et al. [30] investigated the optimal prevention method of PNI during SVCI and found that HPSD radiofrequency energy application (50 W, 7 s), only on SVC points where pacing stimulated the phrenic nerve, never resulted in PNI. Therefore, HPSD energy delivery may represent an optimal PNI prevention maneuver, due to the shallower and wider lesions as compared to the standard radiofrequency ablation. No sinus node injury has been reported in the meta-analysis, although previous study reported this complication in 1.1% of cases [31]. As shown by Dong et al. [24], electroanatomical mapping-guided SVCI allows the localization of the sinus node with successful ablation without any sinus node damage.

Overall, the current meta-analysis shows that the addition of SVCI to PVI seems to provide lower rates of AF recurrences at follow-up as compared with PVI alone, without increasing complication risk. However, our results must be interpreted with caution. Indeed, the number of complications was too limited to draw solid conclusions about safety of SVCI. Furthermore, though we found a significant reduction in terms of AF recurrences in PAF patients, the small number of included patients and the heterogeneity in ablation strategies found among studies entail confirmation of our results in further larger, well-designed RCTs to fully address the effectiveness and safety profile of SVCI.