Ablation therapies for paroxysmal atrial fibrillation: A systematic review and patient-level network meta-analysis

ABSTRACT

Introduction: Despite promising trials, catheter ablation is still regarded as an adjunct to antiarrhythmic drugs (AADs) in the treatment of paroxysmal atrial fibrillation (PAF). This study aimed to compare the effectiveness of various ablation therapies and AADs.

Method: Randomised controlled trials or propensity score-matched studies comparing atrial tachyarrhythmia recurrence among any combination of ablation modalities or AAD were retrieved. Kaplan-Meier curves and risk tables for this outcome were graphically reconstructed to extract patient-level data. Frequentist network meta-analysis (NMA) using derived hazard ratios (HRs), as well as 2 restricted mean survival time (RMST) NMAs, were conducted. Treatment strategies were ranked using P-scores.

Results: Across 24 studies comparing 6 ablation therapies (5,132 patients), Frequentist NMA-derived HRs of atrial fibrillation recurrence compared to AAD were 0.35 (95% confidence interval [CI]=0.25–0.48) for cryoballoon ablation (CBA), 0.34 (95% CI=0.25–0.47) for radiofrequency ablation (RFA), 0.14 (95% CI=0.07–0.30) for combined CBA and RFA, 0.20 (95% CI=0.10–0.41) for hot-balloon ablation, 0.43 (95% CI=0.15–1.26) for laser-balloon ablation (LBA), and 0.33 (95% CI=0.18–0.62) for pulmonary vein ablation catheter. RMST-based NMAs similarly showed significant benefit of all ablation therapies over AAD. The combination of CBA + RFA showed promising long-term superiority over CBA and RFA, while LBA showed favourable short-term efficacy.

Conclusion: The advantage of ablation therapies over AAD in preventing atrial tachyarrhythmia recurrence suggests that ablation should be considered as the first-line treatment for PAF in patients fit for the procedure. The promising nature of several specific therapies warrants further trials to elicit their long-term efficacy and perform a cost-benefit analysis.

Atrial fibrillation (AF) is a pertinent health condition that is seeing a sustained rise in global incidence and prevalence.¹ In particular, paroxysmal AF (PAF), despite its transient nature, is associated with a slow but steady progression to persistent AF² and increased mortality compared to patients without AF.³ Furthermore, patients with PAF have a similar rate of ischaemic stroke as patients with persistent AF; it is thus important to address PAF to prevent future cardiovascular mortality and morbidity.⁴

Medical therapy with antiarrhythmic drugs (AADs) has long been the mainstay of PAF therapy.⁵ Catheter ablation is an alternative upfront treatment with a class IIa recommendation⁵ and can also be performed if AAD therapy fails; the class I recommendation only applies to patients with heart failure with reduced ejection fraction.⁶ The commonly used ablation techniques in practice are radiofrequency ablation (RFA)⁷ and cryoballoon ablation (CBA); others include laser balloon ablation (LBA), hot balloon ablation (HBA), and pulmonary vein ablation catheter (PVAC). Recent trials have shown significant benefit for catheter ablation as a first-line therapy for PAF, and meta-analyses have indeed advocated for ablation to replace AADs in treating PAF.^8,9 However, they were limited by low numbers of included trials, or grouping all ablation modalities as a single arm. Similarly, there are meta-analyses comparing CBA and RFA, but included studies do not usually involve an AAD arm for reference.^10,11

Given the diversity of modalities and fragmented nature of research into the field of ablation therapies for PAF, an all-encompassing synthesis of the subject using evidence from high-quality trials is sorely needed. This can be addressed by a network meta-analysis (NMA), which is an effective method to pool effect sizes across multiple studies and modalities.¹² As randomised controlled trials (RCTs) in this field are limited in number, propensity-score matched studies (PSMs) were also the sought for pooling of individual patient data (IPD). PSMs have been shown to be empirically equivalent to RCTs in generating unbiased estimates of the efficacy of treatment, while eliminating confounding factors and biases to a large extent.¹³

Hence, the aim of this IPD-NMA was to evaluate the effectiveness of various ablation therapies or medical therapy in PAF patients from RCTs or PSMs, with atrial tachyarrhythmia recurrence as the primary outcome.

METHOD

Literature search

This systematic review and IPD-NMA were performed in line with the PRISMA Guidelines for IPD Systematic Reviews. The study design and review protocol were registered with PROSPERO (CRD42022313230).

An electronic literature search from inception to 29 October 2021 was performed on PubMed, EMBASE, Web of Science and the Cochrane Controlled Register of Trials (CENTRAL), without language restrictions. The search strategy included the concepts of PAF and various ablation modalities (Supplementary Table S1 in the online version of this article). Bibliographies of included studies were screened to ensure the inclusion of all relevant studies. The abstract and full-text review were conducted by 2 independent investigators; conflicts were resolved after discussion among all the authors in this paper.

The inclusion criteria were: RCTs or PSMs comparing any combination of ablation modalities or comparing ablation modalities against AADs in patients with paroxysmal atrial fibrillation; and provision of Kaplan-Meier curves with risk tables for atrial tachyarrhythmia recurrence (or complete study follow-up if risk tables were absent). If multiple publications of the same study were retrieved, the most recent and informative publication was included. Studies comparing ablation versus AAD without specifying the exact ablation modality were excluded, as were case reports, case series, reviews and conference abstracts.

Data extraction from included studies was conducted by 2 independent investigators using a standardised data collection template with predefined data fields including study design, patient demographics and outcomes. RCTs were assessed for risk of bias using the Cochrane Risk of Bias (RoB 2) tool for RCTs, and PSMs were assessed using the Newcastle-Ottawa Scale.

Extraction of individual patient data

Given the rapid advancements in ablation therapies in recent years, more precise methods are needed to quantify comparisons among different modalities. Hence, a graphical reconstructive algorithm¹⁴ grounded in analysis outlined by Guyot et al.¹⁵ was employed to attain IPD for survival from atrial tachyarrhythmia recurrence, using Kaplan-Meier curves from included studies. An IPD meta-analysis is recognised as the gold standard approach for evidence synthesis.^16,17

Network meta-analysis

First, a Frequentist NMA was conducted to compare the treatments of CBA, RFA, CBA + RFA, PVAC, LBA, HBA and AAD. Hazard ratio (HR) estimates and their 95% confidence intervals (95% CI) for each IPD study were pooled together in a 2-stage NMA. Treatment strategies were ranked using P-scores, with higher P-scores corresponding to greater efficacy.

As Cox-based models were used to derive HRs for the aforementioned analysis, the proportional hazards assumption was verified. If the assumption was violated, restricted mean survival time (RMST) for each treatment was also analysed^18-20 using 2 measures—RMST mean difference (RMST-D) and RMST ratio (RMST-R)—which were pooled under respective Frequentist NMAs. As RMST does not utilise HRs and is applicable to nonlinear covariate relationships, it acts as a further sensitivity analysis of the first NMA. Heterogeneity was assessed via the I² statistic, and considered low, moderate, or considerable for I² values <40%, 40–75%, and >75%, respectively.²¹ Cochran’s Q was also used to assess for heterogeneity and inconsistency.

Due to the inclusion of non-randomised studies in the form of PSM in this study, sensitivity analysis was also performed, involving a Frequentist NMA with data from RCTs only.

A further NMA was also conducted by pooling ablation procedural times in studies that compared 2 or more ablation procedures. Outcomes were assessed as mean differences (MD), with RFA as a base comparator.

All analyses were conducted in R version 4.1.2, with P<0.05 regarded as indicating statistical significance. Full details of the statistical analyses are provided in the online Supplementary Methods. Ethics approval was not required as this study makes use of publicly available data.

RESULTS

Study selection

The search strategy retrieved 800 studies, of which 461 duplicates were removed; the remaining 339 studies were screened by title and abstract. Thirty-eight studies were identified for full-text review, and a final 24 studies (18 RCTs and 6 PSMs) comprising 5,132 patients were included (Fig. 1).

Fig. 1. PRISMA flowchart of included studies

Study characteristics

Nine studies compared CBA with RFA,^22-31 3 compared CBA with AAD,^32-34 5 compared RFA with AAD,^35-392 compared LBA with CBA,^40,41 2 compared PVAC with RFA,^42,43 1 compared HBA with CBA,⁴⁴ and 1 compared HBA with AAD.⁴⁵ The remaining study was a 3-way comparison of CBA, RFA and a combination of both (CBA + RFA)46 (Table 1). A predominantly Caucasian, male population was involved, and the incidence of diabetes mellitus was under 25% (online Supplementary Table S2). The average duration of AF ranged from 0.4–8.6 years. Baseline beta-blocker use ranged from 5.8–67%. Only 1 study allowed AAD regimens to be newly initiated in the treatment arm.³⁵ In studies with an AAD arm, the most common AADs were flecainide and propafenone. Second-generation CBA was the most common subtype of CBA used in most studies, while studies with an RFA arm were mostly a mix of contact force and non-contact force subtypes. A blanking period, wherein AF recurrence after index ablation was not counted as treatment failure, was implemented in almost all studies, typically spanning 90 days. Adverse events are shown in the online Supplementary Table S3. There were no gross differences in serious adverse events between any ablation modality and AAD, or between ablation modalities, although studies were not likely powered to detect such differences. All studies were deemed to have a low risk of bias (online Supplementary Table S4).

Table 1. Characteristics of included studies

Pairwise comparisons

Three or more studies reported direct comparisons for the following PAF treatments: CBA versus AAD, RFA versus AAD, and CBA versus RFA. IPD-derived one-stage Kaplan-Meier comparisons for the primary outcome among 3 studies showed a significant benefit of CBA over AAD (shared-frailty HR=0.44, 95% CI=0.35–0.56, P<0.001) (Fig. 2A). RFA was also superior to AAD (shared-frailty HR=0.26, 95% CI=0.20–0.33, contact-force P<0.001) across 5 studies (Fig. 2B). There was no significant difference in the primary outcome between CBA and RFA (shared-frailty HR=0.91, 95% CI=0.80–1.04, P=0.174) across 10 studies (online Supplementary Fig. S1). Two-stage comparisons were concordant with one-stage comparisons (online Supplementary Figs. S2–4).

Frequentist network meta-analysis

The combined Kaplan-Meier plot of all studies is shown in Fig. 3 and the network plot of all included trials is shown in online Supplementary Fig. S5.

Derived HRs of atrial tachyarrhythmia recurrence compared to AAD were 0.35 (95% CI=0.25–0.48) for CBA, 0.34 (95% CI=0.25–0.47) for RFA, 0.14 (95% CI=0.07–0.30) for CBA + RFA, 0.20 (95% CI=0.10–0.41) for HBA, 0.43 (95% CI=0.15–1.26) for LBA, and 0.33 (95% CI=0.18–0.62) for PVAC (Table 2 and online Supplementary Fig. S6).

Hence, all treatments except LBA showed significant benefits compared to AAD for the primary outcome. Furthermore, CBA + RFA was significantly favoured over CBA (HR=0.41, 95% CI=0.20–0.83) and RFA (HR=0.41, 95% CI=0.21–0.83) but not against other modalities. Tests of heterogeneity (I²=64%, Cochran’s Q=52.8, P<0.001) indicated moderate between-study heterogeneity and justified the use of a random-effects model. P-scores ranked CBA + RFA as the best modality, followed by HBA (Table 3).

Restricted mean survival time network meta-analysis

When the pairwise comparison with the largest number of direct studies (CBA versus RFA) was analysed, the proportional hazards assumption was violated (online Supplementary Fig. S7). Hence, further analysis was performed using RMST-based NMAs. Separate RMST-R and RMST-D analyses both demonstrated that all ablation therapies were superior to AAD (online Supplementary Figs. S8–11, online Supplementary Tables S5 and S6), with RMST-R versus AAD ranging from 1.33–2.11 and RMST-D versus AAD ranging from 0.15–0.23 year. In the RMST-R analysis, CBA + RFA was significantly superior to all therapies except HBA. P-scores ranked CBA + RFA as the best therapy (Table 2).

Sensitivity analysis

When restricted to the 18 included RCTs only, the hazard ratio-based Frequentist NMA showed similar results to the entire cohort (online Supplementary Fig. S12, Supplementary Table S7). Moderate heterogeneity was observed (I²=75%, Cochran’s Q=51.3, P<0.001).

Network meta-analysis of procedural time

The NMA of procedural time across various therapies was found to have high heterogeneity and inconsistency between studies (I²=92%, Cochran’s Q=120, P<0.001). PVAC was ranked as the fastest procedure, followed by CBA; both had a significantly faster procedural time than RFA (online Supplementary Table S8, Supplementary Fig. S13). CBA + RFA had a significantly longer procedural time than all other ablation therapies.

DISCUSSION

Since the identification of focal origins of atrial ectopic beats and their successful ablation using radiofrequency energy,⁷ subsequent modalities have aimed to achieve the same outcome via different methods of energy delivery. Alongside technical advancements such as contact force ablation, which provides real-time feedback on tissue contact, and remote magnetic navigation, which allows mapping via stereotaxis, ablation has become increasingly safer and more efficacious.⁴⁷

Accordingly, a meta-analysis of the relative efficacies of CBA versus RFA or ablation versus AAD has been previously performed.¹¹ However, this can only present fragmented viewpoints of the larger picture of ablation therapies for PAF. A recent NMA by Kukendrarajah et al.⁴⁸ compared CBA, RFA, LBA and PVAC across 14 RCTs, and concluded that the efficacy of non-RFA technologies was comparable to that of RFA. However, AAD was not included as a baseline comparator. Messori et al.⁴⁹ performed an RMST-based NMA between CBA, RFA and AAD, but only 5 RCTs were included in the analysis. Hence, there remained a role to synthesise all modalities from high-quality studies using IPD, which is recognised as the gold standard approach for evidence synthesis.¹⁶

This IPD-NMA was congruent with previous results in showing the significance of CBA and RFA over AAD in reducing AF recurrence. Pairwise IPD comparisons, which represent more robust analyses compared to pooled study-level aggregate data, showed a large divergence in Kaplan-Meier curves for both CBA and RFA compared to AAD, suggesting a sustained long-term benefit. The certainty of their benefit is further strengthened by indirect comparisons due to the large number of trials comparing CBA to RFA. Altogether, the clinically significant benefit of commonly used ablation therapies across all 3 NMA models, along with a similar safety and adverse event profile to AAD, strengthens the case for upfront ablation therapy as a suitable alternative for patients who can tolerate the procedure. This NMA also provides Level 1A evidence for these comparisons, in support of prior observational studies that showed the benefit of ablation in reducing overall mortality and stroke,^50,51 and reinforcing current European Society of Cardiology guidelines,⁵ which recommend both ablation and AAD as first-line treatment modalities for PAF depending on patient choice.

Moreover, the significance of these results can also be seen with CABANA, which is the largest RCT of upfront ablation versus AAD to date (2,204 patients in total).⁵² Although only 43% of the CABANA cohort had paroxysmal AF (the rest had persistent or long-standing persistent AF), significant reductions in AF recurrence were seen (HR=0.52, 95% CI=0.45–0.60, P<0.001). No differences in mortality, death or cardiovascular hospitalisation were noted, but a lower-than-expected mortality rate in the AAD group may have contributed to the lower precision of the effect size estimate. CABANA was excluded from our analysis as the exact breakdown of RFA and CBA was not specified; however, severe adverse effect profiles were favourable with a 0.8% incidence of cardiac tamponade.

Within the modalities evaluate in this NMA, CBA and RFA showed similar outcomes in both pairwise and NMA comparisons. Previous RCTs have noted the relative ease of performing CBA compared to RFA, with CBA requiring more extensive fluoroscopic guidance.²³ This comparison is further complicated by the presence of newer technologies (contact force sensing and second-generation cryoballoon catheters), which were not well stratified within included studies to conduct a subgroup analysis. Contact force-guided catheters offer lower radiation exposure, procedural times, and cardiac perforation rates, but their association with other safety outcomes and clinical efficacy remains unclear.⁵³ Second-generation cryoballoons markedly decrease AF recurrence compared to the first generation.⁵⁴ A multicentre comparison of contact-force-guided RFA versus second-generation CBA found no significant difference in 18-month atrial arrhythmia freedom, with no periprocedural deaths in either group,⁵⁵ indicating that both are similarly safe and effective. CBA + RFA was only performed in one included study,⁴⁶ involving RFA followed by 2 CBA freezes in a single procedure. The reported 5-year freedom from AF recurrence of 57%, and gross difference in Kaplan-Meier curves versus other techniques (Fig. 3), make it a frontrunner in this field. This was quantitatively supported by HR-based and RMST-R analyses, which found CBA + RFA significantly superior to CBA or RFA alone. CBA + RFA was associated with significantly fewer pulmonary vein reconnections (PVR) than CBA or RFA. Moreover, the different patterns of PVR observed in CBA and RFA—inferiorly for most of the former, superiorly for most of the latter⁵⁶—are postulated to account for the synergy of combining both in a single procedure. Nonetheless, the NMA of procedural time found it significantly more time-consuming than all other ablation modalities. Accordingly, a more in-depth cost-benefit and safety-profile analysis of this technique is necessary, and more studies should be conducted in view of the current evidence for this technique being based solely on one RCT.

Studies of HBA only followed patients for up to 1 year. This lack of long-term follow-up may have led to the unexpected finding of HBA ranking second by way of P values within all 3 NMA methods. Plotting the results of all studies that used AAD as the control arm (online Supplementary Figs. S14 and S15), the AAD arm of Sohara et al.⁴⁵ had a much greater recurrence of atrial tachyarrhythmia than AAD arms in other studies. All patients in this study were refractory to 1 or more AADs upon enrolment, in contrast to some included studies^34,36 that excluded patients previously treated with an AAD. Combined with the slight variation in AADs used, it is possible that the relatively poor performance of the AAD arm in this study is a random error compounded by the low number of patients (n=43). Hence, the relatively inflated value of the HR for this study, which led to its high P-score ranking, may not be a true reflection of its efficacy. Nonetheless, HBA has been noted to have a more favourable learning curve compared to CBA,⁵⁷ which may render it a favourable starting point for centres without prior specialisation in ablation. The remaining modalities, despite not being widely used, similarly showed significant benefits over AAD. PVAC is a modified form of RFA using a multielectrode, circular, bidirectional catheter; it may offer lower procedural times compared to conventional RFA but is no longer widely used due to concerns over the considerable levels of pulmonary stenosis after treatment.⁵⁸ LBA was significantly favoured over AAD in the RMST-based analyses but not the HR-based analysis, which may be attributed to the presence of only 2 LBA studies and its short follow-up. The visually guided LBA used was similar to RFA in requiring point-by-point ablations, as the aiming beam produces an arc covering only ~30º of a circle. LBA was ranked as the third-best modality by way of P-scores in both RMST analyses. The questionable ranking of HBA as the second-best therapy, as mentioned previously, may point to LBA being a stronger contender for second place. From Kaplan-Meier curves alone (Fig. 3), the trajectory of LBA resembles CBA + RFA—the highest-ranked modality—more than other modalities. A recent meta-analysis of LBA versus CBA suggested a trend towards higher 12-month procedural success for LBA,⁵⁹ but no head-to-head studies of LBA versus AAD have been performed. Hence, further trials with long-term follow-up are needed to fill these gaps in evidence.

Pulsed-field ablation (PFA) has emerged in recent years as a tool for pulmonary vein isolation. Nonetheless, promising outcomes in terms of myocardial tissue specificity and dramatically lower procedural time, combined with a unique safety profile, have been found in single-arm trials.⁶⁰ Although one of the major points in favour of CBA usage is its lower procedural time, PFA may offer a promising alternative in the future. Further RCTs are needed to compare PFA to the myriad of technologies in this review.

Despite the use of IPD reconstruction as a rigorous statistical method that accounts for follow-up and censoring status, a noteworthy limitation was the inability to account for effects exerted by patient-level prognostic covariates on arrhythmia freedom. Moreover, all NMAs are limited by the assumptions of methodological equivalence and the similarity of baseline patient characteristics. Included studies were broadly similar in methodology barring the different study designs (RCT and PSM) which have been shown to be empirically equivalent to a large extent. Conversely, baseline patient characteristics varied considerably across studies—an inevitable limitation for essentially all NMAs. Several potential studies which provided Kaplan-Meier curves without risk tables were excluded, as graphical reconstruction is considerably less accurate without risk tables.¹⁵ Measures of safety outcomes among studies were heterogeneous, and most were not sufficiently powered to detect differences in adverse effects; hence, a NMA of safety outcomes was also not feasible, although CABANA has shown favourable safety outcomes in a more adequately powered cohort.⁵² Outcomes such as stroke, all-cause mortality, and periprocedural complications remain an important avenue for clinical decision-making, and future RCTs of individual ablation modalities should aim to be adequately powered for these outcomes as well. Cost-benefit analysis is also an important consideration, with catheter ablation as a whole shown to be reasonably cost-effective compared to AAD.⁶¹ The results of this NMA, which quantitatively ranks 6 ablation techniques, provide a basis for a comprehensive cost-benefit analysis of individual ablation strategies.

CONCLUSION

In addressing the diverse field of ablation therapies for PAF using findings derived from robust networks of high-quality studies, this NMA found a consistent advantage of ablation therapies over AADs in preventing atrial tachyarrhythmia recurrence. Coupled with the large divergence of Kaplan-Meier curves in the CBA versus AAD, and RFA versus AAD pairwise comparisons, ablation represents a suitable first-line alternative to AAD for PAF in patients who are fit for the procedure. Intriguingly, the combination of CBA + RFA showed promising long-term superiority over conventional techniques, while LBA showed favourable short-term efficacy that may warrant a long-term investigation in future trials.

SUPPLEMENTARY MATERIALS

REFERENCES

1. Kornej J, Börschel CS, Benjamin EJ, et al. Epidemiology of Atrial Fibrillation in the 21st Century. Circ Res 2020;127:4-20.
2. Kerr CR, Humphries KH, Talajic M, et al. Progression to chronic atrial fibrillation after the initial diagnosis of paroxysmal atrial fibrillation: Results from the Canadian Registry of Atrial Fibrillation. Am Heart J 2005;149:489-96.
3. Friberg L, Hammar N, Pettersson H, et al. Increased mortality in paroxysmal atrial fibrillation: report from the Stockholm Cohort-Study of Atrial Fibrillation (SCAF). Eur Heart J 2007;28:2346-53.
4. Friberg L, Hammar N, Rosenqvist M. Stroke in paroxysmal atrial fibrillation: report from the Stockholm Cohort of Atrial Fibrillation. Eur Heart J 2010;31:967-75.
5. Hindricks G, Potpara T, Dagres N, et al. 2020 ESC Guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS): The Task Force for the diagnosis and management of atrial fibrillation of the European Society of Cardiology (ESC) Developed with the special contribution of the European Heart Rhythm Association (EHRA) of the ESC. Eur Heart J 2021;42:373-498.
6. Huang W, Keng FY, Ching CK. Rate or Rhythm Control of Atrial Fibrillation – Pearls for the Internist. Ann Acad Med Singap 2017;46:433-8.
7. Haïssaguerre M, Jaïs P, Shah DC, et al. Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med 1998;339:659-66.
8. Kheiri B, Simpson TF, Przybylowicz R, et al. Ablation Versus Antiarrhythmic Drugs as First-Line Treatment of Paroxysmal Atrial Fibrillation: A Meta-Analysis of Randomized Trials. Circ Arrhythm Electrophysiol 2021;14:e009692.
9. Bonanno C, Paccanaro M, La Vecchia L, et al. Efficacy and safety of catheter ablation versus antiarrhythmic drugs for atrial fibrillation: a meta-analysis of randomized trials. J Cardiovasc Med (Hagerstown) 2010;11:408-18.
10. Murray MI, Arnold A, Younis M, et al. Cryoballoon versus radiofrequency ablation for paroxysmal atrial fibrillation: a meta-analysis of randomized controlled trials. Clin Res Cardiol 2018;107:658-69.
11. Buiatti A, von Olshausen G, Barthel P, et al. Cryoballoon vs. radiofrequency ablation for paroxysmal atrial fibrillation: an updated meta-analysis of randomized and observational studies. Europace 2017;19:378-84.
12. White IR. Network Meta-analysis. Stata J 2015;15:951-85.
13. Austin PC. The use of propensity score methods with survival or time-to-event outcomes: reporting measures of effect similar to those used in randomized experiments. Stat Med 2014;33:1242-58.
14. Liu N, Zhou Y, Lee JJ. IPDfromKM: reconstruct individual patient data from published Kaplan-Meier survival curves. BMC Med Res Methodol 2021;21:111.
15. Guyot P, Ades AE, Ouwens MJNM, et al. Enhanced secondary analysis of survival data: reconstructing the data from published Kaplan-Meier survival curves. BMC Med Res Methodol 2012;12:9.
16. Riley RD, Lambert PC, Abo-Zaid G. Meta-analysis of individual participant data: rationale, conduct, and reporting. BMJ 2010;340:c221.
17. Higgins JPT, Li T, Deeks JJ. Choosing effect measures and computing estimates of effect. In: Higgins JPT, Thomas J, Chandler J, et al. Cochrane Handbook for Systematic Reviews of Interventions version 6.3 (updated February 2022). Cochrane, 2022. www.training.cochrane.org/handbook. Accessed on 1 March, 2022.
18. Royston P, Parmar MKB. Restricted mean survival time: an alternative to the hazard ratio for the design and analysis of randomized trials with a time-to-event outcome. BMC Med Res Methodol 2013;13:152.
19. Kim DH, Uno H, Wei LJ. Restricted Mean Survival Time as a Measure to Interpret Clinical Trial Results. JAMA Cardiol 2017;2:1179-80.
20. Royston P, Parmar MKB. The use of restricted mean survival time to estimate the treatment effect in randomized clinical trials when the proportional hazards assumption is in doubt. Star Med 2011;30:2409-21.
21. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ 2003;327:557-60.
22. Buist TJ, Adiyaman A, Smit JJJ, et al. Arrhythmia-free survival and pulmonary vein reconnection patterns after second-generation cryoballoon and contact-force radiofrequency pulmonary vein isolation Clin Res Cardiol 2018;107:498-506.
23. Kuck KH, Brugada J, Fürnkranz A, et al. Cryoballoon or Radiofrequency Ablation for Paroxysmal Atrial Fibrillation. N Engl J Med 2016;374:2235-45.
24. Larsen JM, Deyell MW, Macle L, et al. Impact of Left Common Pulmonary Veins in the Contact-Force vs. Cryoballoon Atrial Fibrillation Ablation (CIRCA-DOSE) Study. J Cardiovasc Electrophysiol 2020. doi: 10.1111/jce.1465
25. Pak H-N, Park J-W, Yang S-Y, et al. Cryoballoon Versus High-Power, Short-Duration Radiofrequency Ablation for Pulmonary Vein Isolation in Patients With Paroxysmal Atrial Fibrillation. Circ Arrhythm Electrophysiol 2021;14:e010040.
26. Pérez-Castellano N, Fernández-Cavazos R, Moreno J, et al. The COR trial: a randomized study with continuous rhythm monitoring to compare the efficacy of cryoenergy and radiofrequency for pulmonary vein isolation. Heart Rhythm 2014;11:8-14.
27. Ikenouchi T, Nitta J, Nitta G, et al. Propensity-matched comparison of cryoballoon and radiofrequency ablation for atrial fibrillation in elderly patients. Heart Rhythm 2019;16:838-45.
28. Knecht S, Sticherling C, von Felten S, et al. Long-term comparison of cryoballoon and radiofrequency ablation of paroxysmal atrial fibrillation: a propensity score matched analysis. Int J Cardiol 2014;176:645-50.
29. Matta M, Anselmino M, Ferraris F, et al. Cryoballoon vs. radiofrequency contact force ablation for paroxysmal atrial fibrillation: a propensity score analysis. J Cardiovasc Med (Hagerstown) 2018;19:141-7.
30. Tokuda M, Matsuo S, Isogai R, et al. Adenosine testing during cryoballoon ablation and radiofrequency ablation of atrial fibrillation: A propensity score-matched analysis. Heart Rhythm 2016;13:2128-34.
31. Andrade JG, Champagne J, Dubuc M, et al. Cryoballoon or Radiofrequency Ablation for Atrial Fibrillation Assessed by Continuous Monitoring. Circulation 2019;140:1779-88.
32. Andrade JG, Wells GA, Deyell MW, et al. Cryoablation or Drug Therapy for Initial Treatment of Atrial Fibrillation. N Engl J Med 2021;384:305-15.
33. Kuniss M, Pavlovic N, Velagic V, et al. Cryoballoon ablation vs. antiarrhythmic drugs: first-line therapy for patients with paroxysmal atrial fibrillation. Europace 2021;23:1033-41.
34. Wazni OM, Dandamudi G, Sood N, et al. Cryoballoon Ablation as Initial Therapy for Atrial Fibrillation. N Engl J Med 2021;384:316-24.
35. Kuck KH, Lebedev DS, Mikhaylov EN, et al. Catheter ablation or medical therapy to delay progression of atrial fibrillation: the randomized controlled atrial fibrillation progression trial (ATTEST). Europace 2021;23:362-9.
36. Morillo CA, Verma A, Connolly SJ, et al. Radiofrequency ablation vs antiarrhythmic drugs as first-line treatment of paroxysmal atrial fibrillation (RAAFT-2): a randomized trial. JAMA 2014;311:692-700.
37. Pappone C, Vicedomini G, Augello G, et al. Radiofrequency catheter ablation and antiarrhythmic drug therapy: a prospective, randomized, 4-year follow-up trial: the APAF study. Circ Arrhythm Electrophysiol 2011;4:808-14.
38. Wazni OM, Marrouche NF, Martin DO, et al. Radiofrequency ablation vs antiarrhythmic drugs as first-line treatment of symptomatic atrial fibrillation: a randomized trial. JAMA 2005;293:2634-40.
39. Wilber DJ, Pappone C, Neuzil P, et al. Comparison of antiarrhythmic drug therapy and radiofrequency catheter ablation in patients with paroxysmal atrial fibrillation: a randomized controlled trial. JAMA 2010;303:333-40.
40. Chun JKR, Bordignon S, Last J, et al. Cryoballoon Versus Laserballoon: Insights From the First Prospective Randomized Balloon Trial in Catheter Ablation of Atrial Fibrillation. Circ Arrhythm Electrophysiol 2021;14:e009294.
41. Yano M, Egami Y, Ukita K, et al. Impact of myocardial injury and inflammation due to ablation on the short-term and mid-term outcomes: Cryoballoon versus laser balloon ablation Int J Cardiol 2021;338:102-8.
42. Gal P, Aarntzen AE, Smit JJ, et al. Conventional radiofrequency catheter ablation compared to multi-electrode ablation for atrial fibrillation. Int J Cardiol 2014;176:891-5.
43. McCready J, Chow AW, Lowe MD, et al. Safety and efficacy of multipolar pulmonary vein ablation catheter vs. irrigated radiofrequency ablation for paroxysmal atrial fibrillation: a randomized multicentre trial. Europace 2014;16:1145-53.
44. Suruga K, Suenari K, Nakano T, et al. Comparison between cryoballoon and hot balloon ablation in patients with paroxysmal atrial fibrillation. J Interv Card Electrophysiol 2022;64:281-290.
45. Sohara H, Ohe T, Okumura K, et al. HotBalloon Ablation of the Pulmonary Veins for Paroxysmal AF: A Multicenter Randomized Trial in Japan. J Am Coll Cardiol 2016;68:2747-57.
46. Ang R, Hunter RJ, Lim WY, et al. Long Term Outcome and Pulmonary Vein Reconnection of Patients Undergoing Cryoablation and/or Radiofrequency Ablation: Results from The Cryo Versus RF Trial. J Atr Fibrillation 2018;11:2072.
47. Cappato R, Calkins H, Chen S-A, et al. Updated Worldwide Survey on the Methods, Efficacy, and Safety of Catheter Ablation for Human Atrial Fibrillation. Circ Arrhythm Electrophysiol 2010;3:32-8.
48. Kukendrarajah K, Papageorgiou N, Jewell P, et al. Systematic review and network meta-analysis of atrial fibrillation percutaneous catheter ablation technologies using randomized controlled trials. J Cardiovasc Electrophysiol 2020;31:2192-205.
49. Messori A, Bartoli L, Ferracane E, et al. Medical therapy, radiofrequency ablation or cryoballoon ablation as first-line treatment for paroxysmal atrial fibrillation: interpreting efficacy through restricted mean survival time and network meta-analysis. Rev Cardiovasc Med 2021;22:557-61.
50. Friberg L, Tabrizi F, Englund A. Catheter ablation for atrial fibrillation is associated with lower incidence of stroke and death: data from Swedish health registries. Eur Heart J 2016;37:2478-87.
51. Bunch TJ, May HT, Bair TL, et al. Atrial fibrillation ablation patients have long-term stroke rates similar to patients without atrial fibrillation regardless of CHADS2 score. Heart Rhythm 2013;10:1272-7.
52. Packer DL, Mark DB, Robb RA, et al. Effect of Catheter Ablation vs Antiarrhythmic Drug Therapy on Mortality, Stroke, Bleeding, and Cardiac Arrest Among Patients With Atrial Fibrillation: The CABANA Randomized Clinical Trial. JAMA 2019;321:1261-74.
53. Ariyarathna N, Kumar S, Thomas SP, et al. Role of Contact Force Sensing in Catheter Ablation of Cardiac Arrhythmias: Evolution or History Repeating Itself JACC Clin Electrophysiol 2018;4:707-23.
54. Giovanni GD, Wauters K, Chierchia G-B, et al. One-Year Follow-Up After Single Procedure Cryoballoon Ablation: A Comparison Between the First and Second Generation Balloon. J Cardiovasc Electrophysiol 2014;25:834-9.
55. Squara F, Zhao A, Marijon E, et al. Comparison between radiofrequency with contact force-sensing and second-generation cryoballoon for paroxysmal atrial fibrillation catheter ablation: a multicentre European evaluation. Europace 2015;17:718-24.
56. Kühne M, Suter Y, Altmann D, et al. Cryoballoon versus radiofrequency catheter ablation of paroxysmal atrial fibrillation: biomarkers of myocardial injury, recurrence rates, and pulmonary vein reconnection patterns. Heart Rhythm 2010;7:1770-6.
57. Yamasaki H, Aonuma K, Shinoda Y, et al. Initial Result of Antrum Pulmonary Vein Isolation Using the Radiofrequency Hot-Balloon Catheter With Single-Shot Technique. JACC Clin Electrophysiol 2019;5:354-63.
58. Compier MG, Leong DP, Marsan NA, et al. Duty-cycled bipolar/unipolar radiofrequency ablation for symptomatic atrial fibrillation induces significant pulmonary vein narrowing at long-term follow-up. Europace 2013;15:690-6.
59. Wei Y, Zhang N, Jin Q, et al. Comparison of efficacy and safety of laser balloon and cryoballoon ablation for atrial fibrillation—a meta-analysis J Interv Card Electrophysiol 2019;54:237-45.
60. Reddy Vivek Y, Neuzil P, Koruth Jacob S, et al. Pulsed Field Ablation for Pulmonary Vein Isolation in Atrial Fibrillation. J Am Coll Cardiol 2019;74:315-26.
61. Aronsson M, Walfridsson H, Janzon M, et al. The cost-effectiveness of radiofrequency catheter ablation as first-line treatment for paroxysmal atrial fibrillation: results from a MANTRA-PAF substudy. Europace 2015;17:48-55.