Improving outcomes after catheter ablation of atrial fibrillation: better patient selection, better procedure, or both?

Abstract

This editorial refers to ‘Temporal trends in atrial fibrillation recurrence rates after ablation between 2005 and 2014: a nationwide Danish cohort study’^†, by J.L. Pallisgaard et al., on page 442.

It has now been 20 years since Haïssaguerre et al. demonstrated that catheter ablation of pulmonary vein triggers could eliminate atrial fibrillation (AF).¹ Since then, catheter ablation has been adopted worldwide as an important therapy for patients with symptomatic, medically refractory AF.² Catheter ablation is more effective than antiarrhythmic drug therapy in patients with recurrent paroxysmal and persistent AF. Despite important developments in catheter ablation technology—including irrigated catheters, electroanatomic mapping systems, and novel ablation approaches—in addition to pulmonary vein isolation, steerable sheaths, or contact-force-sensing catheters, outcomes remain imperfect. For example, in the FIRE and ICE trial, only 63% of patients with paroxysmal AF remained free from AF at 1 year follow-up without antiarrhythmic medication,³ and 59% of patients with persistent AF remained free from AF at 18 months with or without antiarrhythmic medication in the STAR-AF II (Substrate and Trigger Ablation for Reduction of Atrial Fibrillation Trial Part II) trial.⁴ At first sight, these success rates are not much different from success rates reported a decade^{5  ,  6} or two decades¹ earlier.

Yet there is some cause for optimism. Rhythm outcomes following ablation have improved from 52% freedom from AF at 1 year without antiarrhythmic medication between 1995–2002 to 70% between 2003–06.^{5  ,  6} Importantly, a report in this issue of the European Heart Journal adds a systematic analysis of temporal trends of AF ablation success rates in an unselected population undergoing a first AF ablation in clinical practice. Pallisgaard et al. analyzed the risk of recurrent AF following a first AF ablation over time in an inclusive, unselected nationwide Danish cohort.⁷ Using the Danish national register, they identified 5392 patients undergoing first-time catheter ablation for AF between 2005–14. The authors assessed the incidence of recurrent AF after ablation in the Danish registry using a surrogate outcome for freedom from AF without antiarrhythmic drug therapy that was constructed from hospitalization for AF, cardioversion for AF, a repeat AF ablation procedure, and the use of an antiarrhythmic medication ≥90 days after ablation. After adjusting for age, sex, duration of AF, chronic kidney disease, hypertension, heart failure, diabetes, vascular disease, and cardioversion in the year before the ablation procedure, Pallisgaard et al. found that the risk of recurrent AF at 1 year decreased from 45% between 2005–06 to 31% between 2013–14 [adjusted odds ratio (OR) 0.57, 95% confidence interval (CI) 0.47–0.68, P < 0.001)]. Consistent with these results, the risk of repeat ablation also decreased over this same time period (OR 0.61, 95% CI 0.45–0.79).

The data from the Danish registry evoke several important questions. First, are these data generalizable? Are they similar to other nationwide experiences? When the Danish data are compared to other international experiences, the changes over time appear to be consistent. For example, the 45% incidence of recurrent AF in 2005–06 in the Danish registry is quite close to the 48% reported from the first worldwide survey⁶ and the 31% in 2005–06 in the Danish registry is also quite close to the 30% reported in the updated worldwide survey.⁵ The rates of repeat ablation (a more easily measured and reproducible outcome parameter) also appear to be comparable to other large cohorts. The 17% rate of repeat ablation in Denmark is remarkably similar to that observed in nationwide US data (17%)⁸ and a little lower than reported in West and North Europe in the European Society of Cardiology AF Ablation Pilot Registry (23%).⁹

Second, does the increased success rate reflect a change in patient characteristics over time? A review of the patient characteristics shows that the patients undergoing AF ablation indeed changed over time. Compared to 2005–06, patients ablated in 2013–14 had lower rates of amiodarone use, were less likely to have an AF duration ≥2 years, and were less likely to have undergone a cardioversion procedure prior to ablation. This is in line with the trend to make AF ablation accessible earlier and to more patients.² Moreover, when the risks of recurrent AF were examined in several high-risk subgroups such as those with cardioversion before ablation (i.e. persistent AF), and those with heart failure, diabetes, or chronic kidney disease, there was no change in recurrence rates over time.

Third, was the decreased risk of recurrent AF over time also due to improved ablation technique, or only due to the selection of healthier patients ablated at an earlier point in their disease course?¹⁰ The answer is ‘probably both’. While AF ablation was more often performed as an alternative to a second antiarrhythmic drug, or even as first-line antiarrhythmic therapy over the study period, in line with changing recommendations,² several advances in ablation technique were introduced into clinical practice in Europe during this same time period (see above). A sceptic will be quick to point out that the differences in the patients over time may largely explain the changes in outcomes over time. However, the growing confidence and experience in the ablation centres, a better standardization of ablation technique, and improved technology have equally important effects. Recent work has focused on improving patient selection and forecasting risk of recurrence in order to enable better medical decision-making for both patients and physicians. The CAAP-AF score was developed to predict recurrent AF after ablation. The acronym-based score gives points for the presence of coronary artery disease, left atrial diameter, age, persistent or long-standing persistent AF, number of prior failed antiarrhythmic drugs, and female sex.¹¹ APPLE is the acronym for another, comparable score [one point for age >65 years, persistent AF, impaired eGFR (<60 ml/min/1.73 m²), LA diameter ≥43 mm, EF < 50 %)].¹² These clinical risk scores and other developments such as pre-procedural atrial fibrosis staging should help improve patient selection and subsequent outcomes (Take home figure). Importantly, we may need to identify clinical markers for the major drivers of AF in different patients and define different types of AF patients based on their major pathophysiology, to allow selection of the best ablation target leading to stratified or personalized AF therapy.^{13  ,  14}

Take home figure.

Observed (and future) improvement in outcomes following atrial fibrillation ablation are likely the result of several important processes, including improvements in pre-procedure risk stratification, procedural innovation, and improved ablation techniques. These advances in both patient selection and ablation techniques improve patient response rates. Furthermore, these improvements also offer the potential to facilitate our understanding of AF heterogeneity and how we can tailor or personalize AF ablation to improve outcomes even further.

There are some additional and important limitations to consider when evaluating the national Danish data. First, the recurrence endpoint used in the analysis, while useful for estimating clinically important recurrences, is an imperfect surrogate endpoint. The composite endpoint included antiarrhythmic therapy beyond 90 days, which does not always indicate a recurrence, though a sensitivity analysis excluding these endpoints yielded similar results. Additionally, many patients with recurrent AF do not have symptoms and therefore may not be hospitalized, placed on antiarrhythmic therapy, or undergo repeat ablation.¹³ The authors were also unable to account for several important risk factors for recurrent AF, including left atrial size, sleep disordered breathing, or obesity.

What are the implications of this report? First and perhaps most importantly, they provide valid data showing increased overall ablation success over time. However, it is also important to note that success rates did not improve in patients with comorbidities such as diabetes, chronic kidney disease, heart failure, and long-standing AF. Clinicians and investigators should focus on these high-risk patients in order to improve outcomes following ablation. This is particularly true in patients with heart failure, given the challenges associated with medical therapy in heart failure complicated by AF.¹⁵ While we await the outcome of prognostic trials of AF ablation and early AF therapy¹⁰, the current benefits of catheter ablation of AF include decreased cardiovascular hospitalization in randomized clinical trials and deceased healthcare resource utilization in clinical practice.¹⁶ Total health care costs can be reduced even further with improved efficacy of ablation. In US practice, total healthcare costs are 46% higher in patients who require repeat ablation.⁸ Thus, the improvements in the Danish registry also have implications for the economic value of catheter ablation to healthcare systems.

The present analysis, in an unselected and inclusive nationwide cohort of patients undergoing AF ablation, demonstrates that the risk of recurrent AF after ablation has improved over time. Whether this improvement is due to better patient selection, better procedures and techniques, or both, is unknown. However, the quest to improve outcomes after AF ablation should continue to include efforts to improve both patient selection and the procedure itself. Focus on either of these targets in isolation is a fool’s errand.

Funding

This work was partially supported by European Union [grant agreement No 633196 (CATCH ME) to P.K.], the British Heart Foundation (FS/13/43/30324 to P.K.), and the Leducq Foundation to P.K.

Conflict of interest: J.P.P. receives funding for clinical research from Abbott Medical, ARCA biopharma, Boston Scientific, Gilead, and Janssen Pharmaceuticals, and serves as a consultant to Allergan, GlaxoSmithKline, Johnson & Johnson, Medtronic, and Spectranetics. P.K. receives research support from the Medical Research Council (UK), the German Centre for Cardiovascular Research, from several drug and device companies active in AF, and has received honoraria from several such companies. P.K. is listed as inventor on two patents held by the University of Birmingham (Atrial Fibrillation Therapy WO 2015140571 and Markers for Atrial Fibrillation WO 2016012783).