Recurrent Atrial Fibrillation After Ablation: Can Telomere Length Identify Patients Who Are Young at Heart?

Elizabeth S. Kaufman MD


J Cardiovasc Electrophysiol. 2019;30(7):1125-1126. 

Symptomatic atrial fibrillation (AF) is highly prevalent, and this prevalence is expected to increase as the population ages. Many patients experience continued symptoms despite management with atrioventricular nodal blocking agents and antiarrhythmic medication. In addition to being only moderately effective, antiarrhythmic medication is often poorly tolerated and can be associated with both proarrhythmic and systemic side effects. For this reason, ablation has emerged as an attractive treatment strategy. However, ablation carries a risk of serious complications, and there is a substantial risk of recurrence of AF after ablation.[1] The decision whether to proceed to ablation depends, in part, on the perceived likelihood of success. It would be useful for patients and their physicians to have better tools for predicting the success of this procedure.

The study by Su et al[2] in the current issue of Journal of Cardiac Electrophysiology potentially adds another tool for predicting the success of AF ablation. The investigators hypothesized that shortened leukocyte telomere length (LTL), a surrogate for aging and a marker previously related to the risk of developing AF,[3] would be associated with recurrent AF after radiofrequency ablation. They studied 282 consecutive patients with paroxysmal (65%) or persistent (35%) AF refractory to antiarrhythmic drugs.

Paroxysmal AF patients underwent pulmonary vein isolation, with additional lesions for those still in AF after pulmonary vein isolation. Persistent AF patients underwent pulmonary vein isolation and received an additional lesion set. Patients in sinus rhythm after completion of the procedure (which could include cardioversion) and attempted AF induction were considered successfully ablated and eligible for this study. Antiarrhythmic drugs could be used during the 3 months after ablation. Recurrent AF was defined as 30 seconds or more of AF after the 3 month blanking period, documented by ECG or Holter monitor, both of which were obtained routinely at 3, 6, 12, and 18 months after ablation and additionally if symptoms occurred.

Of the 282 patients, 277 patients completed the follow-up and formed the basis for the final data analysis. In this relatively healthy population, only 6% had left ventricular dysfunction and 28% had recurrent AF. Univariate Cox proportional hazard model identified LTL, age, diagnosis-to-ablation time, NT-proBNP, and CHADS2-VASc score as associated with recurrent AF. No significant association was found for multiple other factors, including body mass index, diabetes mellitus, left ventricular dysfunction, left atrial anteroposterior diameter, or high-sensitivity C-reactive protein. Only LTL and diagnosis-to-ablation time remained independent risk markers after multivariable adjustment.

Telomeres are sequences of base pairs at the ends of chromosomes that maintain chromosome stability. However, with each cell division, these chromosome ends fail to replicate completely. Loss of telomere length with aging is highly variable and seems to be accelerated by inflammation.[4,5] Conversely, antioxidants, a healthy diet, exercise, and stress reduction may help reverse the process of telomere shortening.[6] Telomere shortening can cause essential genes to become inactivated, which can lead to apoptosis and predispose to disease states. LTL shortening is associated not only with chronological aging but also with inflammation, oxidative stress, obesity, obstructive sleep apnea, type 2 diabetes, and cardiovascular disease.[6–8] As known risk factors for AF include age, obesity, sleep apnea, diabetes, and heart failure,[9] it should not come as a surprise that shortened LTL is a predictor of recurrent AF. LTL might function as a composite of the other predictors of recurrence. In fact, since telomere shortening can lead to apoptosis, telomere shortening itself might be a mediator of the atrial remodeling that promotes AF.

The other important finding of Su's study was that diagnosis-to-ablation time predicted AF recurrence, whereas neither a diagnosis of persistent (as opposed to paroxysmal) AF nor increased left atrial dimension predicted recurrence. This would suggest that pro-AF remodeling occurs that is independent of atrial dilatation and independent of ongoing continuous AF. It would also support the use of ablation earlier in the management of both paroxysmal and persistent AF. Although diagnosis-to-ablation time has been described previously as a predictor of AF recurrence,[10] the confirmation in the current study reinforces the importance of this measure.

As the investigators acknowledge, this study did not utilize long-term continuous (implantable) monitoring; consequently, some episodes of recurrent AF might have been missed. However, symptomatic AF or AF detected by Holter monitors at 3, 6, 12, and 18 months were at least diagnostic of clinically relevant or persistent AF. A further limitation was that the ablation lesion set varied among patients in this observational study. Nevertheless, the procedures used were consistent with clinical practice, so the results are likely relevant to a wider population.

In summary, the study by Su et al advances our knowledge in several ways. First, it supports the use of LTL as a clinical tool for predicting AF recurrence after ablation. Second, it reinforces the notion that diagnosis-to-ablation time should be minimized to achieve successful outcomes. Perhaps most interestingly, the study provides insight into the role telomere shortening might play in the development of substrate for AF. Like any good research, Su's study raises important questions that should inspire further research.