Superior vena cava isolation is associated with improved outcomes in atrial fibrillation ablation

Abstract

BACKGROUND

Pulmonary vein (PV) isolation is the mainstay in atrial fibrillation (AF) ablation. However, additional arrhythmogenic foci seem to contribute to AF initiation and maintenance. A great proportion of those non-PV foci have been reported to be located in the superior vena cava (SVC).

AIM

To investigate the effectiveness of SVC as an adjunctive therapy to PV isolation.

METHODS

We performed a meta-analysis of MEDLINE and CENTRAL. Inclusion criteria were cohort studies with a control group or randomized clinical trials, comparing patients undergoing AF ablation without additional SVC isolation to those receiving ablation with concurrent SVC isolation, in effects of freedom from atrial tachycardia.

RESULTS

A total of 10 studies, incorporating 2176 patients, were included. The majority of the patients (91.5%) expressed paroxysmal AF. The additional SVC isolation strategy in patients undergoing AF ablation was more effective than the non-SVC isolation strategy [odds ratio (OR) = 0.71; 95% confidence interval (CI): 0.55-0.92]. In a subgroup analysis, radiofrequency ablation demonstrated effectiveness (OR = 0.69; 95%CI: 0.53-0.92). Conversely, the use of cryoablation did not alter clinical outcomes (OR = 0.69; 95%CI: 0.13-3.61). In a distinct subgroup analysis, SVC isolation guided by induction of SVC-originating AF triggers - vs no isolation when no AF-inducing triggers were observed - yielded no superiority (OR = 0.73; 95%CI: 0.46-1.16).

CONCLUSION

The outcomes of AF ablation are favorable when additional SVC isolation is conducted. In a tailored care approach, radiofrequency energy should be preferred. Periprocedural induction of SVC-originating AF triggers, either by isoproterenol infusion or burst atrial pacing, should not be used as a criterion for performing SVC isolation. These findings support integrating SVC isolation into individualized ablation strategies to optimize patient outcomes.

Key Words: Atrial fibrillation; Catheter ablation; Pulmonary veins; Superior vena cava; Non-pulmonary vein triggers

Core Tip: In patients with superior vena cava (SVC) potentials, additional SVC isolation is accompanied by reduced rates of tachycardia recurrence during atrial fibrillation ablation. In a tailored care approach, radiofrequency energy has proven better results compared to cryoablation-mediated SVC isolation. The strategy of periprocedural induction of SVC-originating atrial fibrillation should not be used as a criterion for deciding SVC isolation. These findings support integrating SVC isolation into individualized ablation strategies to optimize patient outcomes.

INTRODUCTION

Atrial fibrillation (AF) is the most frequently encountered arrhythmia in clinical practice[1]. Its association with heart failure, ischemic stroke, and transient ischemic attack is supported by robust evidence[2]. Contemporary data demonstrate the superiority of rhythm control over rate control, in terms of reduced symptom burden and improved quality of life[3]. In the context of rhythm control, therapeutic options include antiarrhythmic drugs and catheter ablation[4]. Pivotal trials support catheter ablation as a first-line strategy for rhythm control due to AF recurrence prevention, reduction of AF burden, and improved quality of life, along with comparable risk of adverse events with antiarrhythmic drug treatment[5,6]. Pulmonary veins (PV) isolation constitutes the cornerstone of AF catheter ablation, ever since the recognition of the PVs as key triggers in the initiation of the arrhythmia[7]. However, additional arrhythmogenic foci have emerged as triggers of the arrhythmia, including the superior vena cava (SVC), the left atrial posterior wall, the crista terminalis, the interatrial septum, and the left atrial appendage[8].

The SVC is considered a major origin of AF triggers, accounting for 25%-40% of non-PV triggers[9], which is attributed to the heterogeneity of SVC myocardial sleeves, containing embryonic sinus venosus tissue, prone to increased automaticity, as well as triggered activity and shorter effective refractory period[10]. The latter is linked to atrial cardiomyopathy and may be manifested as SVC-right atrial junction atrial tachycardia[11-13]. This meta-analysis aims to address the gap in clinical evidence regarding the clinical impact of SVC isolation, on top of PV isolation, in terms of atrial tachyarrhythmia recurrence, along with procedural complications and fluoroscopic-procedural times, by incorporating both randomized clinical trials (RCTs) and observational studies.

MATERIALS AND METHODS

The present meta-analysis was performed according to the Preferred Reporting Items for Systematic reviews and Meta-Analysis[14]. The protocol to conduct this meta-analysis has already been published in PROSPERO.

Search strategy

We systematically searched MEDLINE (through PubMed) and the Cochrane Central Register of Controlled Trials (CENTRAL) to identify studies that evaluated the effectiveness of SVC isolation in addition to PV isolation, in patients undergoing AF ablation, until August 2025. Basic keywords used in the search string were “atrial fibrillation”, “ablation”, and “superior vena cava”. The search syntax can be found in the Online Resource. All references from selected studies were retrieved and manually reviewed according to the snowball effect. In addition, the archives of major cardiological conferences were searched, as well as ClinicalTrials.gov, for relevant ongoing studies. PROSPERO was also searched to identify possible similar systematic reviews in progress to avoid duplication with our study. No language restrictions were applied.

Eligibility and exclusion criteria

We included randomized clinical trials and observational studies comparing the outcomes among patients with AF undergoing ablation with an additional SVC isolation strategy and patients not undergoing additional SVC isolation. All studies should provide sufficient data regarding the ablation approach and outcomes during follow-up. The outcome of interest was freedom from atrial tachyarrhythmias. Observational studies without a control group were excluded.

Study selection

All search results were imported into a reference management software (EndNote 20). All duplicates were removed, and three reviewers (Botis M, Tsiachris D, and Doundoulakis I) independently screened titles and abstracts and perused full texts for eligible studies. A fourth reviewer (Antoniou CK) was consulted to resolve any discordance regarding study eligibility. All reasons for exclusion at the stage of full-text eligibility were recorded.

Data extraction

Two reviewers (Tsiachris D and Doundoulakis I) independently extracted data regarding study design and study outcomes on a structured spreadsheet. A pilot test was performed before initiation to ensure coherence between the review authors. Any disagreement was resolved by consensus. All data were extracted from full-texts, summary tables, and figures or Supplementary material using standard methods as proposed by the Cochrane Collaboration. The authors conducted an analysis to obtain any missing data relevant to the analysis. Data items that were extracted included: Title and first author, year of publication, country/region of participant recruitment, study design, total number of participants, patients’ characteristics [mean age, gender, Congestive Heart failure, Age ≥ 75, Diabetes, Stroke, Vascular disease, Age 65-74, Sex category (CHADSVASC) score, Hypertension, Abnormal renal/Liver function, Stroke, Bleeding, Labile INR, Elderly, Drugs/alcohol (HAS BLED) score, type of AF, left atrium diameter], ablation techniques, and follow-up data. Outcomes during follow-up were recorded as the number of events and/or effect measures whenever provided, along with their corresponding standard errors.

Quality assessment

The methodological quality of the included studies was evaluated by two independent reviewers (Botis M and Tsiachris D) using the Cochrane Collaboration risk of bias tool for randomized controlled trials (RoB 2) and the ROBINS-I (Risk of Bias in Nonrandomized Studies of Interventions) tool for non-randomized trials for intervention[15,16]. A third reviewer (Doundoulakis I) resolved any disagreements. Visualization of the quality assessment results was enabled with the use of the “robvis” tool (available from: https://mcguinlu.shinyapps.io/robvis/)[17].

Statistical analysis

Continuous variables are expressed as mean ± SD, or median and 25%-75% interquartile range. For studies reporting medians and interquartile ranges, we applied the Wan method or Hozo’s correction to convert these estimates into mean ± SD, when necessary. Summary estimates of categorical variables as odds ratios (OR) with 95% confidence intervals (CI). DerSimonian-Laird random-effects model meta-analysis was chosen due to anticipated clinical and methodological heterogeneity between the studies. Pooled results of the meta-analysis were visualized in a forest plot along with the 95%CIs. Heterogeneity between the studies was assessed using Cochran’s Q test and I² statistic for standard thresholds (I²: 25% or lower for low heterogeneity, 26%-50% for moderate heterogeneity, and 50% for high heterogeneity). A two-sided alpha level of 0.05 was considered statistically significant. A prespecified subgroup analysis was also performed. Subgroups were defined according to the source of energy implemented. Publication bias was assessed using a funnel plot and Egger’s regression test with a significance level of 0.1. All analyses were performed using the meta and metafor packages in the R project for Statistical Computing (2021.09.0 Build 351).

RESULTS

Search results

The search strategy yielded 1097 results, after the removal of duplicates. Following the initial screening phase, 65 full-text studies were screened for eligibility, and 55 were excluded for various reasons. A total of ten studies were included in this systematic review, consisting of six RCTs of 1056 patients and four observational studies of 1120 patients. The study selection process can be seen in Figure 1.

Figure 1  Preferred Reporting Items for Systematic Reviews and Meta-Analyses flowchart of the study selection process.

Study and patient characteristics

Among the 10 studies, 6 were randomized controlled trials, and 4 were observational. The exact randomization flowchart for each study is described in Table 1[10,18-26]. Among the observational studies, 3 were prospective, and one was retrospective. Two studies were multicenter, and the rest were single-center. The included studies originated from seven countries. Eight studies utilized radiofrequency ablation, and cryoablation was implemented in two studies (Table 1). The total number of patients included was 2176, with 69.8% being male. The patients were followed up for a mean of 29.9 months. The mean Congestive Heart failure, Age ≥ 75, Diabetes, Stroke, Vascular disease, Age 65-74, Sex category (CHADSVASC) score was 1.79 ± 1.2. The mean Hypertension, Abnormal renal/Liver function, Stroke, Bleeding, Labile INR, Elderly, Drugs/alcohol (HAS BLED) Score was reported solely by Dong et al[24] and was 1. Paroxysmal AF was present in 1990 (91.5%) patients, while 88 (4.0%) patients suffered from persistent, and 98 (4.5%) patients from long-standing persistent AF. Detailed patient characteristics can be found in Table 2.

Table 1 Design and main characteristics of the eligible studies.

Ref.	Type of study	Decision to ablate	Isolation technique	Primary outcome
Wang et al[18], 2008	Single-center, randomized	Randomization, prior to EP study and catheter ablation	Radiofrequency ablation	Atrial tachycardia recurrence
Corrado et al[19], 2010	Single-center, randomized	Randomization, without investigating SVC triggers with provocative maneuvers; if SVC spontaneous potentials were not recorded, no SVC electrical isolation was conducted	Radiofrequency ablation	Maintenance of sinus rhythm without antiarrhythmic drugs
Da Costa et al[20], 2015	Single-center, randomized	Randomization was conducted at the time of the procedure, if the mapping catheter above the right atrium-SVC junction revealed active electrical potentials in the SVC, under sinus rhythm, or under stimulation	Radiofrequency ablation	Freedom from atrial tachycardia and atrial fibrillation
Muto et al[21], 2007	Single-center, observational, prospective	SVC isolation was performed when spontaneous AF originating from the SVC was observed, with reproducibility via isoproterenol infusion and burst atrial pacing	Radiofrequency ablation	Occurrence of electrical connection recovery after segmental ostial isolation in patients with recurrent atrial fibrillation for SVC vs PV
Higuchi et al[10], 2010	Single-center, observational, prospective	SVC isolation was performed if AF triggers originated from the SVC, after AF induction with high-frequency pacing and intravenous isoproterenol infusion	Radiofrequency ablation	Identification of structural and electrophysiologic differences between the SVC of patients with and without SVC triggering of SVC (length of sleeve, voltage)
Takigawa et al[22], 2017	Single-center, observational, prospective	SVC isolation was performed if sustained or non-sustained AF was reproducibly initiated from SVC foci, after isoproterenol infusion or rapid atrial pacing	Radiofrequency ablation	Atrial fibrillation recurrence
Overeinder et al[23], 2021	Single center, observational-retrospective	SVC isolation was performed if electrical activity was documented in the SVC prior to ablation	Cryoballoon ablation	Freedom from atrial tachycardia
Dong et al[24], 2024	Multi-center, randomized	Before randomization, isoproterenol or adenosine triphosphate infusion and rapid burst pacing were used to reveal SVC AF triggers; thirty patients exhibited SVC-triggered AF and underwent SVC electrical isolation; one hundred patients did not exhibit SVC-originated AF triggers and were randomized to PV isolation vs PV isolation and SVC electrical isolation; in the overall analysis, the 80 patients who underwent SVC isolation were compared with the 50 patients who did not undergo SVC isolation; in the subgroup analysis of studies where SVC isolation was conducted after induction of SVC-originating AF triggers with isoproterenol infusion or burst pacing, 30 patients who exhibited SVC-induced AF after isoproterenol infusion or burst pacing were compared with 50 patients who did not exhibit SVC-induced AF and did not undergo SVC electrical isolation	Radiofrequency ablation	Freedom from any documented atrial tachycardia
Castro-Urda et al[25], 2025	Single-center, randomized	The decision to electrically isolate the SVC was taken according to the presence of spontaneous SVC potentials	Cryoballoon ablation	Freedom from atrial fibrillation/atrial flutter/atrial tachycardia
Shen et al[26], 2025	Multi-center, randomized	Randomization to PV isolation plus SVC isolation or solely PV isolation. SVC isolation was conducted in the presence of SVC potentials	Radiofrequency ablation	Freedom from atrial arrhythmias

SVC: Superior vena cava; PV: Pulmonary vein; AF: Atrial fibrillation.

Table 2 Baseline characteristics of the patients included.

Ref.	Country	Total number of patients	Follow up (months)	Atrial fibrillation type	Age1	Left atrium diameter (mm)	Gender (male)	CHA2DS2-VASC/CHADS2 score
Wang et al[18], 2008	China	106	4 ± 2	100% paroxysmal	66.0 ± 8.8	32.85 ± 8.96	52.8%	NR
Corrado et al[19], 2010	Italy	320	12	46% paroxysmal; 23% persistent; 31% long-standing persistent AF	56 ± 4.25	45.76 ± 0.94	67.8%	NR
Da Costa et al[20], 2015	France	100	15 ± 8	100% paroxysmal	56 ± 9	42 ± 2	83%	0.9 ± 1
Muto et al[21], 2007	Japan	95	6	100% paroxysmal	58.47 ± 11	35 ± 5	54.7%	NR
Higuchi et al[10], 2010	Japan	60	12	76.67% paroxysmal; 23.33% persistent	59.2 ± 10	38 ± 4.67	76.67%	NR
Takigawa et al[22], 2017	Japan	865	53.5 ± 39.1	100% paroxysmal	61 ± 10	36.62 ± 5.14	77.46%	0.8 ± 1 (CHADSC score)
Overeinder et al[23], 2021	Belgium	100	12	100% paroxysmal	55.3 ± 5.84	32.85 ± 8.96	68%	1 ± 1
Dong et al[24], 2024	China, Singapore	130	12	100% paroxysmal	58.1 ± 4.7	37.85 ± 3.67	72.0%	1 ± 1
Shen et al[26], 2025	China	302	20	100% paroxysmal	64 (56.0-70.0)	NR	54.6%	2.4 ± 1

¹mean ± SD, median, and interquartile range.

AF: Atrial fibrillation; NR: Not reported.

Clinical outcomes

Patients who underwent additional SVC isolation had lower rates of atrial tachyarrhythmia recurrence during follow-up when compared to the strategy of non-additional SVC isolation (OR = 0.71, 95%CI: 0.55-0.92) (Figure 2A). No heterogeneity was observed among the studies (I² = 0%, P = 0.498). Acute SVC isolation was achieved in 95.5% of patients within the designated subgroups. A trend towards increased rates of complications was noted in the additional SVC isolation group, however it did not reach statistical significance (OR = 2.6, 95%CI: 1.00-6.90) (Supplementary Figure 1). In a distinct analysis incorporating only phrenic nerve injury, the SVC isolation strategy conferred a marginally increased rate of adverse events (OR = 4.10, 95%CI: 1.01-16.63) (Supplementary Figure 2). A detailed table including the most important major complications, namely, is included in the Supplementary Table 1. Procedural duration and fluoroscopy time were prolonged in the SVC isolation group (Supplementary Figures 3 and 4). The sensitivity analysis, incorporating a fixed-effects model, yielded qualitatively comparable results in the overall population, in terms of atrial tachyarrhythmia recurrence (Supplementary Figure 5). The percentage of redo procedures, as well as the reconnection rates, are depicted in Table 3.

Figure 2 Forrest plot. A: Forrest plot of atrial tachyarrhythmia recurrence between superior vena cava (SVC) isolation + pulmonary vein (PV) isolation vs PV isolation-only patients; B: Forest plot of atrial tachyarrhythmia recurrence between SVC isolation + PV isolation vs PV isolation-only, utilizing radiofrequency energy; C: Forest plot of atrial tachyarrhythmia recurrence between SVC isolation + PV isolation vs PV isolation-only, among patients who underwent cryoablation. SVC: Superior vena cava; PVI: Pulmonary vein isolation; OR: Odds ratio; CI: Confidence intervals.

Table 3 Percentage of redo procedures and reconnection rates.

Ref.	Percent of repeat procedure in the SVC group	Any pulmonary vein reconnection percentage	SVC reconnection percentage	Percentage of repeat procedure in the non-SVC group	Any pulmonary reconnection percentage
Wang et al[18], 2008	16%	100%	0%	16.7%	77.8%
Corrado et al[19], 2010	NR	NR	NR	6.88%	NR
Muto et al[21], 2007	26.7%	NR	25%	31.3%	NR
Higuchi et al[10], 2010	0%			31.3%	100%
Takigawa et al[22], 2017	26.3%	60%	53.3%	30.1%	80.3%
Overeinder et al[23], 2021	10%	8%	0%	NR	NR
Castro-Urda et al[25], 2025	16.7%	77.7%	66.7%	20.4%	100%
Shen et al[26], 2025	NR			NR

SVC: Superior vena cava; NR: Not reported.

Subgroup analysis, according to the source of energy used

The outcomes were stratified according to the source of energy used: In the subgroup of radiofrequency ablation, the SVC electrical isolation strategy was associated with lower rates of atrial tachyarrhythmia recurrence during follow-up (OR = 0.69, 95%CI: 0.53-0.92, I²: 0%) (Figure 2B). In contrast, in studies where cryoablation was implemented, recurrence of atrial tachycardia was not significantly reduced (OR = 0.69, 95%CI: 0.13-3.61, I²: 82%) (Figure 2C).

Prognostic association based on SVC isolation after periprocedural induction of AF

The strategy of SVC isolation guided by the induction of AF-triggering ectopic beats after burst pacing or isoproterenol infusion was distinctively tested, compared with no isolation when no AF-inducing triggers were observed. No superiority was demonstrated (OR = 0.73, 95%CI: 0.46-1.16; I²: 0%) (Supplementary Figure 6).

Prognostic association according to racial group

A distinct analysis was made according to the racial origin of the participants. Asian patients demonstrated benefit (OR = 0.71, 95%CI: 0.51-0.99; I² = 0%), while the effect was not reproduced in the Caucasian population (OR = 0.69, 95%CI: 0.38-1.28; I²: 0%) (Supplementary Figures 6 and 7).

Risk of bias

Five RCTs were assessed for the risk of bias with the Cochrane Collaboration risk of bias tool for randomized controlled trials (RoB 2) tool, and four observational studies with the ROBINS-I (Risk of Bias in Nonrandomized Studies of Interventions) tool. In the RCTs, no bias arose from missing outcomes or the measurement of the outcome. One study was downgraded due to potential bias in the randomization process, and deviations from the intended intervention and one study was downgraded due to the bias or deviation from the intended intervention. Visualization of the risk of bias assessment is presented in the Supplementary Figures 8 and 9.

Publication bias and grading of evidence

Publication biases were assessed with a contour-enhanced funnel plot and Egger’s linear regression for the outcome of freedom from atrial tachycardia. No visual signs of bias were observed for the outcome of interest (Supplementary Figure 10), and Egger’s test was not statistically significant (P = 0.965).

DISCUSSION

To the best of our knowledge, this is the first meta-analysis to evaluate the efficacy of SVC electrical isolation in the context of AF ablation, stratified according to radiofrequency vs cryoballoon implementation. Among 2176 patients, where non-SVC isolation strategy was compared with additional SVC isolation, several interesting findings were noted: (1) Additional SVC isolation reduces the risk of recurrent atrial tachyarrhythmia; (2) Radiofrequency ablation is effective in terms of SVC isolation strategy; (3) Cryoballoon implementation is not associated with decreased rates of atrial tachyarrhythmia recurrence; and (4) Additional SVC isolation is associated with increased procedural and fluoroscopy time.

Ever since the PVs were established as the mainstay target of every AF ablation procedure, additional targets have been explored to further enhance procedural efficiency. From an embryologic standpoint, SVC seems an ideal substrate, since it contains cardiac musculature connected to the right atrium[27]. The segments of SVC where SVC potentials can be recorded are referred to as the SVC sleeves, and a long SVC sleeve (> 30 mm) or a large SVC potential within the sleeve (> 1 mV) are strong predictor of SVC AF foci[10]. SVC cardiomyocytes were found to have enhanced automaticity and afterdepolarization properties[28], which account for both spontaneous SVC potentials and SVC-originated AF triggers. Observational studies demonstrated that 25%-50% of non-PV triggers originate from the SVC[9,10]. Current guidelines underline that PV isolation is the cornerstone in AF ablation, while the clinical utility of further ablation strategies is yet to be determined[4]. SVC isolation is currently recognized as an adjunctive target, however, it remains an area of uncertainty[29].

Previous meta-analyses have explored the effect of SVC isolation vs PV isolation, with neutral results in the overall population[30,31]. The superior efficacy of radiofrequency ablation compared to cryoballoon ablation may be attributed to the heightened concern for phrenic nerve palsy associated with cryoballoon use, which often leads to more conservative energy application. However, the high heterogeneity in the cryoballon ablation group warrants cautious interpretation of the findings and may reflect insufficient data rather than a lack of clinical benefit. What is more, the provocation of SVC triggers through isoproterenol infusion or pacing may not be of benefit, even though it is technically feasible, particularly when radiofrequency ablation is implemented well. This finding may warrant significant implications for future procedural workflow planning and therapeutic decisions. The difference in terms of clinical benefit between Asian populations and Caucasian populations should be interpreted with caution, since fewer studies included the Caucasian population. Further research is needed.

Supplemental SVC isolation does not seem to be associated with an increased risk of adverse events. Da Costa et al[20] reported two phrenic nerve injuries in the SVC-isolation group. Muto et al[21] reported one case of transient phrenic nerve paralysis. Higuchi et al[10] stated that transient right diaphragmatic paralysis was noted in one patient during SVC isolation. Castro-Urda et al[25] reported two vascular complications, ten instances of transient phrenic nerve paralysis, and nine cases of temporary sinus bradycardia. Shen et al[26] reported one case of transient phrenic nerve paralysis and two cases of periprocedural tamponade in the SVC isolation group. High-output pacing can be utilized during SVC isolation to capture the phrenic nerve and timely recognize an impending phrenic nerve injury.

Limitations

The key methodological issue of our meta-analysis is the different study designs and ablation strategies implemented. Moreover, varying follow-up periods could have contributed to variations in outcomes. It is worth noting that the cryoballoon group suffered from increased heterogeneity. Moreover, the relatively restricted number of studies may limit the statistical power to detect smaller effects, especially in the cryoballoon subgroup. Variations in operators’ experience and preferences could also impact the effectiveness of PV isolation and SVC isolation. Although no pulsed field ablation studies were included, we do not believe, given recent evidence of similar efficacy[32], that it would have affected findings - if anything, reduced risk of phrenic nerve palsy, compared with previous energy sources, could allow for a lower threshold for performing SVC isolation[33,34]. In addition, data were lacking to compare the rates of redo procedures between the SVC isolation strategy and the non-SVC isolation strategy. From a methodological standpoint, the external validity of the study may be reduced by the inclusion of both randomized and observational studies. However, we sought to adjust by stratifying across different ablation strategies. Finally, our findings should be extrapolated with caution in persistent and long-standing AF patients, since the overwhelming majority of study participants had paroxysmal AF, thereby limiting applicability to more advanced forms of the arrhythmia. Consequently, cautious interpretation of our results is recommended when implementing them into clinical decision-making.

CONCLUSION

In conclusion, in patients with SVC potentials, additional SVC isolation is accompanied by reduced rates of tachycardia recurrence. In a tailored care approach, radiofrequency energy has proven better results compared to cryoablation-mediated SVC isolation. The strategy of periprocedural induction of SVC-originating AF should not be used as a criterion for deciding SVC isolation. These findings support integrating SVC isolation into individualized ablation strategies to optimize patient outcomes.