Atrial fibrillation (AF) is the most common sustained arrhythmia and it has been related to increased morbidity and mortality.1,2 Nowadays, there is a growing interest both in its early detection and in a better discrimination of those patients at a higher risk that would obtain more benefit from anticoagulation to prevent thromboembolic complications. Dual-chamber pacemakers have demonstrated to provide reliable detection of subclinical AF (SCAF), defined as atrial high rate episodes (AHRE), and clinical AF (electrocardiographically [ECG] documented).3 AHRE have been related to a higher risk of both clinical AF and thromboembolic events.4 Previous studies have related atrial pacing (AP) with a reduction in AF recurrences5; however, this relationship is controversial, since other authors suggest an increased risk of AF related to AP.6 Ventricular pacing (VP) has also been related to an increased risk of AF due to different mechanisms.7 Therefore, pacing is related to an increased risk for future clinical AF development.

The aim of this study was to identify if AP and/or VP are related not only to higher rates of clinical AF but also to SCAF, as a previous stage in the natural course of AF, in patients with dual-chamber pacemaker and no prior history of this arrhythmia.

The aim of our study was to evidence a relationship between the percentage of AP and VP and the incidence of SCAF and clinical AF in patients with dual-chamber pacemakers and no history of AF. We could demonstrate an association between AP and the risk of developing SCAF and clinical AF; moreover, AP was shown to be an independent risk factor. VP was also related to an increased risk of SCAF and clinical AF. Moreover, patients with both AP/VP ≥ 50%, had a tendency towards a higher risk of SCAF. The importance of early detection of SCAF and clinical AF is crucial due to its high embolic risk; the results of our study are of high interest because we could use high percentages of AP and VP as early markers of atrial arrhythmias.

We demonstrated a clear association between AP and the risk of subclinical and clinical AF. Patients with AP ≥ 50% had 3 times more risk of developing AHRE and 2 times more risk of developing clinical AF. Previous studies have related AP with a risk of developing AF, but the mechanisms remain unclear.6 AP produces an atrial conduction delay with intra- and interatrial dyssynchrony, that has been related to the development of AF. The alteration of the conduction times could also be related to higher rates of VP, which has also been associated to an increased risk of AF.8 On the other hand, some studies have suggested AP as a mechanism for AF prevention. The ADOPT study proposed an algorithm that increased the pacing rate when the native rhythm emerged. The result was a reduction in AF risk in patients with sick sinus syndrome.5 In ASSERT, continuous overdrive AP did not prevent AF.9 However, some other studies have shown that AP might be detrimental for new AF onset in patients without sinus node disease, in ICD patients10 and in CRT patients.11 Nevertheless, it remains unclear whether this is related to atrial stimulation or to the increased atrial rate, as in the DAVID trial.12 Therefore, there is great controversy about the relationship between AP and AF, and different mechanisms might be involved. These patients sometimes require AP because of a sick sinus syndrome; therefore, they would have a more diseased atrium, more likely to develop AF. Longer baseline PQ interval has been associated with an increased risk of AF8; this probably also reflects a prolonged atrial conduction time associated with atrial fibrosis. To note, there are different reasons why pacing percentage can be increased. In our study, 22% of patients were under betablockers (Table 1), and the main reason for the pacemaker implantation in patients with AP ≥ 50% was sinus node dysfunction (Table 2). A feasible explanation would be that these patients frequently suffer from tachycardia-bradycardia syndrome and so would be more likely to present AF. In this case, AP could be a marker of risk and not the mechanism for developing AF. On the other hand, 26% of the patients in our study had a previous diagnosis of cardiomyopathy; however, none of them had a diagnosis of cardiac amyloidosis or cardiac sarcoidosis that would predispose them to developing AF.

We could also demonstrate an association between VP and the risk of SCAF and AF. Patients with VP ≥ 50% had almost twice the risk of developing AHRE and almost three times more risk of developing clinical AF. The association between VP and AF has been previously suggested.7 There are some studies that have related VP with an increased risk of permanent AF in patients with sinus node disease.13 However, the effects of VP in paroxysmal AF are controversial. Although previous studies suggested that VP was not detrimental in patients with paroxysmal AF, results of long-MinVPACE conclude that VP was related to an increase in AF burden in patients with paroxysmal AF.14 The mechanisms by which VP will increase the risk for AF are based on the loss of atrioventricular synchrony, which is associated with an increase in atrial pressures, enlargement of the left atrium and mitral regurgitation, all of them related to AF. In this sense, dual-chamber pacing will reduce rates of AF in comparison to VP.13 Also, prolonging programmed atrioventricular intervals with the intention to reduce the rates of VP, has been proposed to reduce the incidence of AF in pacemaker patients.

The percentage of VP can be influenced by the device programming. As already mentioned, it is the protocol of our arrhythmia unit to avoid VP by prolonging the atrioventricular interval and using other dedicated algorithms when necessary. Atrioventricular block was the main reason for pacemaker implantation in patients showing VP ≥ 50% (Table 5). Conventional dual chamber pacing is usually indicated for atrioventricular block, for atrioventricular synchrony maintenance. However, sometimes it can result in high percentage of VP leading to ventricular dyssynchrony, which has been linked by previous studies with an increased risk of AF.13

The role of pacing to prevent AF has been widely discussed. AF is usually the consequence of multiple reentrant wavelets or the impossibility of following a 1:1 activation, leading to a fibrillatory conduction.15 AP would decrease the bradycardic episodes as well as the ectopic beats and so it would eliminate the reentry mechanisms for AF. Multisite AP and different atrial locations of pacing have been tested without showing a decrease in AF risk. AAI or DDDR pacing are less related to the development of AF in comparison to VVI pacing.15 Moreover, atrial rate could have a role in addition to that of AP.

Some studies have suggested an increased risk of hospitalization for heart failure related to ventricular dyssynchrony secondary to VP.16 In our study, we could not demonstrate an association between higher rates of AP and VP and clinical events but pacing was related to a higher incidence of AHRE resembling SCAF, with a previously related higher risk of stroke, as a prior stage for future development of clinical AF. The definition of AHRE has been widely discussed; in our study, the definition of AHRE as episodes ≥ 5minutes was based on studies as MOST4or ASSERT9 that showed that these episodes of> 6 and> 5minutes were related to embolic events. On the other hand, a previous study of our group showed that we should be careful with AHRE interpretation at the early implantation period, since they are more frequent and not related to a higher risk of embolism17. However, in our study, although there was some variability in the times from device implantation, it did not influence the outcomes.

The importance of these results lies on the current increasing interest in the early detection of silent AF, both in patients with cardiac implantable electronic devices, as well as in the global population, through new wearable technologies (iWatch, Kardia…). The detection of episodes of AHRE would be the equivalent of detecting an incipient stage of AF in its progressive course, ranging from paroxysmal AF to persistent and permanent AF. The progression of AF in its different stages has been associated with a greater number of clinical events.18 AHRE have also been related to a high risk of stroke and systemic embolism. Despite the demonstrated increased risk of stroke in patients with SCAF, in the absence of results of randomized controlled trials, the presence of AHRE is not currently an indication for anticoagulation. Two important randomized controlled trials (ARTESIA and NOAH-AFNET) are ongoing, and their results may establish a future indication for the anticoagulation of these patients.19,20.

Pacing is related with a higher risk of developing SCAF and clinical AF in patients with dual-chamber pacemaker and no history of previous AF. Our data suggest that patients presenting a high percentage of AP should be closely followed during routine pacemaker check-ups, assessing for AHRE and clinical AF.

No funding to declare.

No.