https://orcid.org/0000-0002-6618-7184
Abstract
Convincing evidence for the efficacy of ablation as first-line therapy in paroxysmal AF (PAF) and its clear superiority to medical therapy for rhythm control in both PAF and persistent AF (PsAF) has generated considerable interest in the optimal timing of ablation. Based on this data, there is a widespread view that the principle of ‘the earlier the better’ should be generally applied. However, the natural history of AF is highly variable and non-linear, and for this reason, it is difficult to be emphatic that all patients are best served by ablation early after their initial AF episodes. Sufficient evidence exists to indicate a conservative approach is reasonable in patients with infrequent and non-progressive episodes (i.e. absence of progressive increase in burden culminating in PsAF) in whom symptoms remain mild and well-controlled. A conservative management phase should be marked by assiduous attention to risk factor modification, changes in frequency and duration of AF episodes, and patient preferences. If and when AF does begin to progress, accumulating evidence indicates that early ablation accompanied by ongoing attention to risk factors provides the best outcomes.
Varied scenarios in the timing of atrial fibrillation (AF) ablation. For some, a wait-and-manage strategy can be employed. For others, early ablation is clearly indicated with the primary goal of symptom reduction. Legend: *Requires watchful waiting and aggressive risk factor management. **After discussion of the uncertain natural history including the possibility that burden may remain low for years. ‡Symptom and AF burden comparable to randomized trials. ‡‡After a detailed discussion of evidence supporting ablation as more successful at improving symptoms. †After detailed discussion of lack of evidence supporting superiority of this approach. AAD, anti-arrhythmic drug; BMT, best medical therapy; DCR, direct current cardioversion; EO, Expert opinion only (no randomized data available); HFrEF, heart failure with reduced ejection fraction; LVEF, left ventricular ejection fraction; PAF, paroxysmal atrial fibrillation; PsAF, persistent atrial fibrillation; QoL, quality of life; RFM, risk factor management; SR, sinus rhythm; Sympt, symptomatic comparable to patients included in randomized trials.
Introduction
In multiple randomized trials, catheter ablation for the management of atrial fibrillation (AF) results in greater freedom from AF recurrence and improved quality of life (QoL) compared with anti-arrhythmic medications.1–6 Convincing evidence for the efficacy of ablation as first-line therapy, at least in paroxysmal AF (PAF), and its clear superiority to medical therapy for rhythm control in both PAF and persistent AF (PsAF) has generated considerable interest in the optimal timing of ablation. This comes at a time where, under the impact of dramatically advancing technology, ablation efficacy has progressively increased as procedural risk has significantly declined.7,8 Nevertheless, the rate of major complications in contemporary studies is not negligible, and this must also be considered in the context of ablation timing.8
Is there then a broadly applicable ‘sweet spot’ beyond which ablation efficacy begins to decline if ablation is deferred for a trial of, or indeed ongoing medical therapy? Does a failure to act early result in reduced ablation efficacy later, and is the patient also exposed to increasing risk of adverse events such as systemic embolism, heart failure, and premature mortality9? Indeed, if ablation is inevitable due to the inexorable progression of AF in the vast majority of medically treated patients, is there not an imperative to act early in all? This concern is particularly significant in the context of evidence that AF may drive progressive and potentially irreversible atrial remodelling.
Recent debates at international meetings with titles such as ‘The earlier the better: AF ablation should be performed as early as possible’ reflect an increasing view in the field that AF ablation should be performed as early as possible after diagnosis in order to control symptoms, prevent progressive remodelling and progression of the AF phenotype from paroxysmal to persistent, and to prevent adverse events.
This has considerable implications, not the least of which pertains to resources: in many geographies, it can be difficult to ablate patients even in the more urgent categories in a timely fashion.
Early ablation
There are clearly circumstances for which there is wide agreement that ablation should be performed expeditiously. These include established symptomatic PAF or PsAF either with failed medical therapy or as first-line therapy (PAF only) when this accords with patient preference, as it frequently does. Both of these scenarios received an advice level of ‘To Do’ in the latest international consensus document.10 The evidence for both was derived from multiple randomized trials.2–6 This advice was further supported by a Class 1, level of evidence A recommendation by the 2023 ACC/AHA AF guideline document for two similarly worded statements11; although the latter recommendation was qualified by the words ‘generally younger and with few comorbidities’. Ablation in these circumstances results in a significant and sustained reduction in AF burden accompanied by improvements in QoL12 and mental health,13 reduction in the need for anti-arrhythmic medication, and reduction in healthcare costs.14
Nevertheless, it should be emphasized that these statements specified that the indication was for the management of AF symptoms and AF burden and for the prevention of AF progression.15 There has as yet been no recommendation for the performance of AF ablation to ameliorate the risk of major adverse cardiovascular events (MACE) with the exception of patients with heart failure. Indeed, while the EAST-AFNET 4 trial recently established that early rhythm control therapy in patients with AF of duration <12 months is associated with a reduction in a primary composite outcome of cardiovascular death, stroke, heart failure hospitalization, or acute coronary syndrome,16 the majority of patients were managed with medical rhythm control. Just 8% at enrolment and 19% by 2 years of follow-up had undergone ablation, indicating that anti-arrhythmic therapy continues to play an important role in many patients early on in the natural history of AF.
The evidence for early AF ablation in the context of heart failure with reduced ejection fraction (HFrEF) is clearly settled. Numerous studies have demonstrated improvement or indeed full recovery of left ventricular function with the resumption of sinus rhythm (SR) and its long-term maintenance. As such, this carries a ‘To Do’ or Class 1 recommendation in both recent documents.10,11 While the most marked improvements are observed in patients without evidence of late gadolinium enhancement on magnetic resonance imaging, even those with myocardial scar can derive significant benefits.17 Furthermore, catheter ablation in HFrEF patients may be associated with reduced mortality even in the context of end-stage heart failure.18–20 Thus the recent Class 1 ACC/AHA recommendation was, ‘In patients who present with a new diagnosis of HFrEF and AF, arrhythmia-induced cardiomyopathy should be suspected, and an early and aggressive approach to AF rhythm control is recommended’.11 This recommendation does leave the door open to cardioversion and medical therapy if it provides effective rhythm control, but ablation will frequently be preferred.
Finally, there is widespread recognition that patients with long-standing PsAF (>12 months) have lower success rates for catheter ablation (compared with PsAF of <12 months duration), frequently requiring multiple procedures and more extensive ablation.21 Thus, for patients with PsAF in whom maintenance of SR cannot be readily achieved, there is a clear imperative to progress to ablation when appropriate.
However, it is much less clear that these indications for early ablation necessarily translate to all patients presenting with AF (Graphical Abstract). Indeed the indication for ‘first-line’ ablation is frequently interpreted as an indication for ‘first-time’ ablation; this is particularly so in the context of diagnosis-to-ablation time (DAT) data suggesting a narrow window of opportunity after apparent first diagnosis.
PAF natural history: is progression to PsAF inevitable and inexorable?
AF is widely considered to be a progressive disease, and in many instances this is true. But is this inevitable, and over what time course?
Over the past few decades, a large amount of information regarding the natural history of AF has been accumulated. A recent systematic review and meta-analysis included 47 studies with 27 266 AF patients who were followed for 105 912 patient-years.22 The study observed an incidence of progression from PAF to non-PAF of 7.1 per 100 patient-years of follow-up (or 7.1%/year). Most striking, however, was the marked variability between rates of reported progression, with variation from as low as 0.8% up to 35.6%. Despite this, the majority of studies clustered around 7–8%, and almost three-quarters of the heterogeneity could be accounted for by variables including age, hypertension (HT), and the duration of follow-up (the longer the follow-up, the lower the rate of progression). The impact of follow-up duration reflects the fact that AF progression does not occur in a linear fashion but also depends on when in their natural history patients are enrolled in a study. Patients referred later in their disease course due to increasing symptoms (and followed from that point) are likely to be a group with a higher incidence of progression.
Several studies have demonstrated an association between AF progression and MACE. In one study of patients with implanted devices who had died, the analysis demonstrated increasing AF burden in the weeks prior to death.23 However, this study did not have access to the prevalence of risk factors or comorbidities and thus could not ascertain causation. In an analysis of the Euro Heart Survey, patients who progressed to PsAF had a higher incidence of MACE. However, these patients also had a higher incidence of major risk factors and comorbidities, and again causation could not be definitively established.24
Studies of AF progression have frequently lacked a detailed characterization of AF burden prior to the study as a potentially important determinant of progression. In addition, a lack of high-density monitoring may result in an underestimation of AF progression. Recently, two studies have looked at the incidence of progression in patients with PAF using implanted loop recorders.15,25 Andrade et al compared the rate of progression from paroxysmal to PsAF in anti-arrhythmic drug (AAD) naïve patients randomized to either ablation or anti-arrhythmic therapy.15 The fact that these patients had not previously been prescribed AAD medications suggests a group very early after AF diagnosis. However, the median period since AF diagnosis in this study was one year (with an interquartile range of up to 4). This emphasis is important as these were mostly not patients early after first diagnosis nor had they had only their first few AF episodes. Patients were highly symptomatic, with a median of 3 (and up to 10) symptomatic episodes per month.2 This study therefore does not provide support for ‘first-time’ ablation. Despite the relatively high AF burden in the study when considering event frequency, the rate of progression to PsAF in the AAD group was only 7.4% at 3 years. It should be emphasized that this is unlikely to be an underestimate, as patients were continuously monitored with implanted loop recorders. As there was a 1.9% incidence of progression in the ablation group, the absolute difference in progression to PsAF after 3 years was 5.5%. Although this landmark study indicates that ablation is not essential to prevent AF progression early in the natural history of AF in the overwhelming majority of PAF patients, it does nevertheless demonstrate the potential for this important positive benefit.
In an observational study, Ngyuen et al. followed 417 PAF patients with implanted loop recorders and a mean time since AF diagnosis of 2.6 years.25 Over a further 2.2 years of follow-up, 8.4% of patients progressed to PsAF (3.8%/year) and a further 3.8% demonstrated increment in PAF burden (defined as >3% increase). In contrast, 11.5% of patients had no AF at all in this 2.2-year period and the vast majority (76.3%) continued to have low burden paroxysms without evidence of progression. In this study, predictors of AF progression included male sex, increased left atrial volume and decreased contractility, moderate mitral regurgitation, waist circumference, and biomarkers including N-terminal pro-B-type natriuretic peptide and markers of inflammation.
Both of these studies indicate that the majority of patients with PAF will not demonstrate progression over the next 2–3 years but that there are predictors of who is more likely to progress.
Indeed, numerous different risk scores have been proposed to predict not only incident AF and AF ablation outcomes but also the likelihood of AF progression.26,27 The common features are generally age, comorbidities, and lifestyle risk factors with a range of biomarkers also being evaluated.25,28,29 For example, the HATCH score (HT, Age >75, TIA/stroke, chronic obstructive pulmonary disease, and heart failure) was previously found to be a moderate predictor of AF progression.24 Groups with higher risk scores require careful attention to risk factor modification, closer follow-up, and early intervention when AF does progress.
‘First-line’ ablation does not equate to ‘first-time’ ablation
An analysis of inclusion criteria, AF duration, and AF burden of the six randomized trials comparing first-line ablation to AAD therapy in PAF indicates that these trials included patients with AF duration largely in excess of 6 months and established evidence of recurrent AF2,6,30–33 (Table 1). Although AF frequency and burden were presented differently in the various trials, in general burden was high and indeed, a significant percentage of patients had undergone prior cardioversion. These studies did not include patients after the first episode of AF.
Inclusion criteria, AF burden and AF duration for randomized studies of first-line ablation vs. anti-arrhythmic therapy
| Study | Inclusion criteria | AF episode no. | AF duration (years) | Prior DCR |
|---|---|---|---|---|
| Andrade 20212 | Non-permanent* AF within the last 24 months:
| 3 symptomatic episodes/month (1–10) | 1 (0–4) | 39% |
| Wazni 20216 | Recurrent self-terminating AF within past 6 months with ECG documentation | NA | 1.3 ± 2.5 | 16% |
| Kuniss 202131 | ≥2 PAF episodes in past 6 months | AF burden 10.9 ± 15.6% | 0.8 ± 2.1 | NA |
| Morillo 201432 | Recurrent PAF ≥4 PAF episodes in past 6 months | 47 ± 98 episodes past 6 months | NA | 43% |
| Cosedis Nielsen 201230 | ≥2 PAF episodes in past 6 months < 7 days ‘considered to be appropriate candidates for ablation’ | 57% patients ≥ 2 episodes/week. 88% patients ≥ 2 episodes/month | NA | 31% |
| Wazni 200533 | At least monthly symptomatic AF episodes for at least 3 months | 13 ± 6 AF episodes in ≥3 months | 5 months | NA |
| Study | Inclusion criteria | AF episode no. | AF duration (years) | Prior DCR |
|---|---|---|---|---|
| Andrade 20212 | Non-permanent* AF within the last 24 months:
| 3 symptomatic episodes/month (1–10) | 1 (0–4) | 39% |
| Wazni 20216 | Recurrent self-terminating AF within past 6 months with ECG documentation | NA | 1.3 ± 2.5 | 16% |
| Kuniss 202131 | ≥2 PAF episodes in past 6 months | AF burden 10.9 ± 15.6% | 0.8 ± 2.1 | NA |
| Morillo 201432 | Recurrent PAF ≥4 PAF episodes in past 6 months | 47 ± 98 episodes past 6 months | NA | 43% |
| Cosedis Nielsen 201230 | ≥2 PAF episodes in past 6 months < 7 days ‘considered to be appropriate candidates for ablation’ | 57% patients ≥ 2 episodes/week. 88% patients ≥ 2 episodes/month | NA | 31% |
| Wazni 200533 | At least monthly symptomatic AF episodes for at least 3 months | 13 ± 6 AF episodes in ≥3 months | 5 months | NA |
Inclusion criteria, AF burden and AF duration for randomized studies of first-line ablation vs. anti-arrhythmic therapy
| Study | Inclusion criteria | AF episode no. | AF duration (years) | Prior DCR |
|---|---|---|---|---|
| Andrade 20212 | Non-permanent* AF within the last 24 months:
| 3 symptomatic episodes/month (1–10) | 1 (0–4) | 39% |
| Wazni 20216 | Recurrent self-terminating AF within past 6 months with ECG documentation | NA | 1.3 ± 2.5 | 16% |
| Kuniss 202131 | ≥2 PAF episodes in past 6 months | AF burden 10.9 ± 15.6% | 0.8 ± 2.1 | NA |
| Morillo 201432 | Recurrent PAF ≥4 PAF episodes in past 6 months | 47 ± 98 episodes past 6 months | NA | 43% |
| Cosedis Nielsen 201230 | ≥2 PAF episodes in past 6 months < 7 days ‘considered to be appropriate candidates for ablation’ | 57% patients ≥ 2 episodes/week. 88% patients ≥ 2 episodes/month | NA | 31% |
| Wazni 200533 | At least monthly symptomatic AF episodes for at least 3 months | 13 ± 6 AF episodes in ≥3 months | 5 months | NA |
| Study | Inclusion criteria | AF episode no. | AF duration (years) | Prior DCR |
|---|---|---|---|---|
| Andrade 20212 | Non-permanent* AF within the last 24 months:
| 3 symptomatic episodes/month (1–10) | 1 (0–4) | 39% |
| Wazni 20216 | Recurrent self-terminating AF within past 6 months with ECG documentation | NA | 1.3 ± 2.5 | 16% |
| Kuniss 202131 | ≥2 PAF episodes in past 6 months | AF burden 10.9 ± 15.6% | 0.8 ± 2.1 | NA |
| Morillo 201432 | Recurrent PAF ≥4 PAF episodes in past 6 months | 47 ± 98 episodes past 6 months | NA | 43% |
| Cosedis Nielsen 201230 | ≥2 PAF episodes in past 6 months < 7 days ‘considered to be appropriate candidates for ablation’ | 57% patients ≥ 2 episodes/week. 88% patients ≥ 2 episodes/month | NA | 31% |
| Wazni 200533 | At least monthly symptomatic AF episodes for at least 3 months | 13 ± 6 AF episodes in ≥3 months | 5 months | NA |
DAT and the imperative to ablate early after first diagnosis
Against the backdrop of possible AF progression, accumulating evidence evaluating outcomes on the basis of the DAT has indicated that the longer you wait to ablate, the less likely it is that ablation will be successful.34
The DAT has been evaluated as a predictor of ablation outcome in a number of retrospective studies and database analyses. A 2020 meta-analysis of six retrospective ablation series suggested that delaying ablation by over 1 year from the time of first diagnosis resulted in a 27% increment in overall risk of AF recurrence.34 However, patient populations in the majority of these series were generally poorly matched, reflecting ablation selection bias.35,36 Patients with longer DAT were frequently older with a higher prevalence of HT, heart failure, and coronary artery disease (CAD), higher CHA2DS2-VASc, and had more comorbidities such as chronic obstructive pulmonary disease, all of which may have significantly contributed to the outcome differences.34
A recent retrospective study compared two cohorts of patients who underwent ablation over a decade apart and observed the most marked effects of the DAT on ablation outcomes in the first 3 years after diagnosis.37 However, the short DAT group was predominantly drawn from the more recent group of patients who underwent ablation with contemporary technology, a fact that no doubt significantly impacted the findings. Numerous other advances in management over that decade may also have contributed to differences in outcomes, such as recognition of the importance of risk factor management (RFM).
A recent claims and encounters database analysis of over 11 000 patients indicated an adjusted risk of AF recurrence of 20% for every 1-year increment in DAT.38 The linear relationship indicated no threshold effect after diagnosis. In striking contrast, a large Danish nationwide registry study of 7705 patients with either paroxysmal or PsAF spanning almost a decade found that virtually all of the impact of DAT on outcomes occurred within the first year.39 The difference in AF recurrence between the first quartile (DAT < 1 year) and the second quartile (DAT 1–1.9 years) was 12% with no further clinically significant decline in recurrence between the second quartile (median DAT, 1.5 years) and the fourth quartile (median DAT, 7.7 years). It seems implausible that from the time of diagnosis of PAF, there is a window of 1 year before atrial remodelling has progressed to such a point that any further remodelling will have minimal impact on ablation outcomes. This study was also confounded by significant differences between quartiles in multiple critical risk factors, including age, CHA2DS2-VASc score, and prevalence of diabetes, HT, CAD, and stroke.
In contrast, data obtained from the recent prospective randomized CAPLA study in patients with PsAF found only a relatively minor impact of the DAT on ablation outcomes (∼2.5% reduction in ablation efficacy per year of DAT).40 In this study, with a few exceptions, the quartiles were well-matched for risk factors and comorbidities. We hypothesize that in a prospective study with standardized entry criteria, most patients will be at a similar time-point in terms of AF disease progression, despite the differences in time taken to reach that point due to variable natural history. Therefore, differences in outcome by quartile of DAT would be minimized. Indeed, atrial mapping in that study indicated no significant differences in electrical remodelling (voltage or conduction) between the four quartiles. The study was subtitled ‘Any time can be a good time to ablate’ to indicate that a long DAT does not necessarily preclude an excellent ablation outcome.
Furthermore, the DAT when evaluated retrospectively, is flawed in a number of important respects. First, it fails to incorporate patient-specific information that determines ablation timing, in particular AF burden and symptom severity. Second, it is most likely highly inaccurate in identifying the first patient episode of AF. Third, it focuses on AF as the major reversible driver of progressive remodelling which may not be true (or wholly true) in many instances, and finally, it does not take into account patients whose benign natural history has meant that they are still doing well without the need for ablation. These latter patients, by definition, are not represented in studies of DAT.
Nevertheless, despite its limitations and the observed variations between studies, studies of the DAT have provided important insight into the adverse impact of delaying ablation in patients with high-burden, ongoing AF. While the DAT may not accurately reflect the time of first AF occurrence in retrospective and database studies, by its nature (i.e. first ECG documentation), it is nevertheless likely to identify a time when AF starts to progress with increasing burden and symptoms leading to ECG documentation. From this point, the imperative to intervene in order to alleviate symptoms, prevent increasing atrial remodelling, and yield the highest likelihood of SR maintenance becomes more pressing.
Finally, a recent small randomized trial indicated no adverse impact of delaying ablation by 12 months on ablation outcomes (maintenance of SR at 12 months post-ablation) in a mixed population of paroxysmal and PsAF patients.41 Importantly, however, the group who underwent delayed ablation were actively treated during the 12 months with modification of anti-arrhythmic therapy and direct current cardioversions (DCRs) when necessary, and with attention to lifestyle modifications. Although a small study, it was adequately powered to detect the very large difference in outcomes that earlier DAT studies suggested occurred after the first DAT year of up to 30%. Furthermore, the outcome curves were superimposed, and the median post-ablation burden was 0% in both early and delayed groups. This study provides some reassurance that in patients with effective medical management, delaying ablation by up to 1 year does not have a major adverse effect on ablation outcomes. However, it is possible that the time cushion of safely delaying ablation is less in persistent AF than in PAF patients, and this should be taken into consideration. It is important to note that this study does not indicate that leaving patients in high-burden or PsAF over a period of a year with inadequate rhythm control is a reasonable strategy. Furthermore, this study does not negate an earlier ablation strategy in patients who prefer that approach.
AF, atrial remodelling and reverse remodelling
Since the pioneering work of Allessie,42 there has been a recognition that AF can drive progressive atrial remodelling, both electrical and structural.42,43 Initially, the AF begets AF phenomenon was based on acute changes in atrial refractoriness, but it subsequently became clear that longer periods of AF result in structural remodeling.43 Further pioneering work from the group of Nattel and others indicated that atrial remodelling may occur in the absence of AF related to factors such as heart failure, HT, and other stimuli, and was termed atrial remodelling ‘of a different sort’.44,45
The association between modifiable risk factors and incident AF has been appreciated for many decades.46–51 In more recent years, mapping studies in humans have demonstrated the impact of these risk factors on atrial substrate in a dose-dependent manner.52–58 The development of conduction slowing, low-voltage zones, and indeed regions of scar, together with complex atrial signals, has been shown for risk factors such as obesity, obstructive sleep apnoea, HT, heart failure, excess alcohol intake and with advancing age. Similar to prediction of progression, more advanced atrial remodelling has been predicted by clinical scoring systems that include a range of risk factors characterized by the APPLE, DR-FLASH, and MB-LATER scores.27,59
Human studies have also indicated more advanced atrial electrical and structural remodelling in patients with PsAF when compared with PAF60,61 indicating not only that AF remodelling progresses over time but also that there is a relationship to AF burden. However, the rate at which AF-driven atrial remodelling progresses and the AF burden required to progress the process are two key issues for which there are limited data. The above studies of DAT, indicating that AF ablation efficacy declines markedly after the first year since diagnosis, imply that AF-driven atrial remodelling progresses rapidly during this period, and further that a year may be long enough for remodelling to become complete and irreversible. This may be plausible for high-burden PAF and for PsAF but it seems improbable that infrequent and relatively brief episodes of AF are responsible for the progression of remodelling that occurs in the weeks and months between episodes. In a cohort of patients with high-burden PAF and implanted loop recorders, Walters et al performed a serial echocardiographic assessment and serial P-wave duration measurements at 4 monthly intervals over 12 months.62 Only those patients with AF burden >10% demonstrated a significant and progressive decline in measures of left atrial strain as a correlate of atrial contractile function and an increment in P-wave duration indicating atrial conduction slowing. In the group with a burden <10%, neither of these parameters changed over that time period. The same study demonstrated that successful catheter ablation can arrest and even reverse this remodelling over a 12-month follow-up period. However, data in this area have been mixed. A similar study performed over a decade ago evaluated atrial electrical remodelling using repeat electrophysiology evaluation, arguably a more sensitive indicator than P-wave measurements.63 This study, performed in an era prior to the implementation of routine RFM, found that atrial electrical remodelling could advance even following successful ablation, thus pointing to factors beyond AF itself that progress remodelling.
Impact of RFM
A series of seminal studies in obese sheep demonstrated that weight gain was a clear causative factor in the development of progressive atrial electrical and structural remodelling.54 Importantly, weight loss in this model demonstrated that, to a large extent, this remodelling may be reversible.64 However, other animal models have indicated that when replacement fibrosis supervenes, remodelling may not be fully reversible if at all.65,66 Small studies in humans have indicated that reverse remodelling does occur with weight loss67 and with the treatment of sleep apnoea.68 More compelling, however, are studies indicating a reduction in AF burden with weight loss and comprehensive RFM.69–73 Most striking is evidence indicating that weight loss and RFM can change the natural history of AF progression.70 In the REVERSE-AF study, maintained weight loss of >10% was associated with a reduction in progression of PAF to PsAF at 5 years to just 3% compared with 48% in those who failed to lose or indeed gained weight. Furthermore, weight loss of >10% was associated with the reversal of the AF phenotype from persistent to paroxysmal in 88% compared with just 26% of patients who did not lose weight. Similarly, in morbidly obese patients, AF gastric sleeve surgery resulted in the reversal of AF phenotype in 71% of patients.73 Furthermore, AF ablation outcomes were significantly improved in morbidly obese patients who had prior bariatric surgery, with recurrence rates similar to a non-obese group.72,74 Similarly, in a randomized study of alcohol abstinence compared with ongoing alcohol intake, those patients who abstained demonstrated a highly significant reduction in AF burden and increment in freedom from AF during the follow-up period.75 There is also emerging evidence that regular physical activity,76 smoking cessation, and treatment of HT all contribute to a decrease in AF recurrences as part of a comprehensive RFM programme.71,77 These data provide compelling evidence that remodelling can be reversible and that in preventing the progression of AF, RFM is a central pillar, standing together with rhythm management.78
Progression point to ablation time
We propose an alternative to the ‘DAT’ in terms of considering the timing of AF ablation, which better reflects the significant variability in AF natural history and progression (Figure 1). The inflection point from when the AF burden starts to progress seems a more appropriate time-point from which to consider ablation. AF progression may be broadly defined by the time-point beyond which increasing AF episode frequency and/or duration begins to impair QoL. Recent data suggests that this may occur at an AF burden of 0.1% or an episode duration of >1 h.79 However, it is probable that this will vary between patients, necessitating individualized and shared decision-making. As previously indicated, we hypothesize that to a large extent, the DAT does best reflect this inflection point, the time at which a diagnostic ECG is most likely to have been obtained in a large proportion of patients. Prior to this inflection point, aggressive RFM (to prevent the associated progressive remodelling), AADs if needed, and ongoing follow-up to ensure patients do not silently progress to PsAF form a reasonable approach. The role of AADs at this stage will often be as a pill in the pocket. When a patient reaches an AF burden or symptom severity necessitating AADs as prophylaxis, a recommendation to undergo ablation would be the preferred approach. The smartwatch or a related device may provide an ideal method for allowing careful follow-up of AF burden in those patients being managed conservatively to ensure early detection of increasing AF burden.80
Progression point to ablation time. Schematic of varied AF progression time course. Some patients have a benign course for years prior to progression, and others progress rapidly from the outset. Top panel: indicative of a patient showing no progression over years. Due to low burden, AF-driven remodeling is likely minimal. Middle panel: Indicative of a patient with early progression to high-burden AF necessitating early ablative intervention. Atrial remodeling likely to progress as a consequence of high burden AF if ablation leading to effective rhythm control is not undertaken. Lower panel: Patient indicative of a long period without progression of burden or symptoms and likely without progressive AF-driven atrial remodeling during this period. The patient then reaches a ‘progression point’ beyond which symptoms and burden progress quite rapidly, necessitating ablative intervention. In each of these three scenarios, atrial remodeling may also be driven by risk factors, and RFM is important across time. (Modified and reproduced from Crowley et al.40 with permission from Elsevier Science and Technology Journals).
In the era of wearables, with the widespread uptake of technologies such as the smartwatch, the time of first diagnosis is likely to shift earlier in the natural history of the condition.81,82 We should not necessarily view this as a point from which the clock of progressive remodelling is inexorably ticking. For many patients, episodes will remain infrequent and of short duration over many years without obvious progression of remodelling.
Conclusions
Convincing evidence for the efficacy of ablation as first-line therapy in PAF and its clear superiority to medical therapy for rhythm control in both PAF and PsAF has generated considerable interest in the optimal timing of ablation. Based on this data, there is a widespread view that the principle of ‘the earlier the better’ should be generally applied. However, the natural history of AF is highly variable and non-linear, and for this reason, it is difficult to be emphatic that all patients are best served by ablation early after their initial AF episodes. Sufficient evidence exists to indicate a conservative approach is reasonable in patients with infrequent and non-progressive episodes (i.e. absence of a progressive increase in burden culminating in PsAF) in whom symptoms remain mild and well-controlled. A conservative management phase should be marked by assiduous attention to risk factor modification, changes in frequency and duration of AF episodes, and patient preferences. If and when AF does begin to progress, accumulating evidence indicates that early ablation accompanied by ongoing attention to risk factors provides the best outcomes.
Supplementary data
Supplementary data are not available at European Heart Journal online.
Declarations
Disclosure of Interest
Dr. Sanders, Dr. Kistler and Dr. Kalman are supported by Clinical Investigator grants from the NHMRC. Dr Sanders reports having served on the advisory board of Medtronic, Abbott Medical, Boston Scientific, CathRx, and PaceMate. Dr Sanders reports that the University of Adelaide has received on his behalf research funding, lecture, and/or consulting fees from Medtronic, Abbott Medical, Boston Scientific, and Becton Dickenson. Dr Kalman has reported receiving research support from Biosense Webster, Boston Scientific, Zoll inc., Abbott inc., and Medtronic. Dr. Kistler reports receiving speaker fees from Abbott medical and advisory fees from Biosense Webster.
Data Availability
No data were generated or analysed for or in support of this paper.
Funding
Nothing to declare.
