Atrial high-rate episodes: a comprehensive review

Summary

Cardiac electronic implantable devices (CIEDs) have the ability to monitor, store and interpret complex arrhythmias, which has generated a new arrhythmic entity: atrial high-rate episodes (AHRE). AHRE are atrial tachyarrhythmias, detected only by CIEDs. They are widely considered a precursor to atrial fibrillation (AF) but can also be represented by other kinds of supraventricular arrhythmias such as atrial flutter or atrial tachycardia. CIED-detected AHRE are associated with an increased risk of stroke, but the risk is significantly lower than the stroke risk of clinical AF. Moreover, there seems to be no temporal correlation between AHRE and thromboembolic events. Because of the current gaps in evidence, the appropriate management of this arrythmia can be challenging. In this review we take into account the epidemiology behind AHRE, predictive factors, clinical impact and management of this arrhythmia.

Atrial fibrillation (AF) is the most common sustained tachyarrhythmia encountered in clinical practice. Because of its high impact on morbidity and mortality, it is also the most studied tachyarrhythmia.¹ Despite the latest progress achieved in managing AF, it still is one of the major causes of stroke, heart failure, sudden death and cardiovascular morbidity.^1-4 Its incidence and prevalence are expected to rise steeply in the following years because of population ageing and the progress achieved in diagnosing asymptomatic episodes.

The recent development of devices capable of long-term continuous monitoring of cardiac rhythm revealed that asymptomatic AF is more frequent than symptomatic AF.^5-8 Despite the obvious difference in quality of life, there is currently no evidence that these two entities have a different risk profile. There are however several prospective ongoing trials that focus on this subject.^9-11

Different options in long-term monitoring of the cardiac rhythm are currently used: handheld electrocardiograph (ECG) monitoring devices, long-term Holter ECG, subcutaneous implantable cardiac monitors (ICM), cardiac implantable electronic devices (CIEDs) and even smartphone ECG applications. A significant advantage of CIEDs when compared to other long-term rhythm-monitoring devices is their ability to continuously monitor cardiac rhythm. Due to their complex algorithms, modern CIEDs are able to store and interpret complex arrythmias, which has resulted in a new arrhythmic entity: atrial high-rate episodes (AHRE).

Definitions

The European Heart Rhythm Association (EHRA) consensus for device-detected subclinical atrial tachyarrhythmias defines AHRE as follows: atrial tachyarrhythmia episodes with an atrial rate of > 190 beats/min (bpm) detected only by CIEDs.¹²

The various definitions used in the literature have generated confusion between AHRE and subclinical atrial tachyarrhythmias (AT). Subclinical (asymptomatic) AT can be detected by a variety of different methods, including external surface-monitoring methods (standard ECGs, Holter monitors or event monitors) and CIEDs [pacemakers, implantable cardioverter defibrillators (ICDs)], while AHRE are detected only by CIEDs. Furthermore, AHRE are widely considered a precursor to AF but can also be represented by other kinds of supraventricular arrythmias such as atrial flutter or atrial tachycardia.¹²

While the impact of AHRE on morbidity and mortality has been proven, there are certain differences when compared to AF. CIED-detected AHRE are associated with a two-fold increase in stroke risk when compared to patients with no AHRE, but the risk is significantly lower than the stroke risk of clinical AF.^13,14 This is the main reason why AHRE must be distinguished from asymptomatic paroxysmal AF diagnosed by surface ECG methods, which usually identify patients with higher burdens of AT. Moreover, several studies have shown that AHRE do not seem to be temporally associated with stroke.^15,16 These two main differences support the idea of two distinct clinical entities.

Incidence and prevalence

The reported incidence of AHRE varies with the definition of AHRE, study design, indication for CIED, presence of AF history, following period and type of device.

Because many CIED-recorded arrhythmias have proven to be inaccurate, the diagnosis of AHRE requires several criteria as well as manual reviewing of the electrogram (EGM). Therefore, a > 190-bpm threshold has been chosen to increase the specificity of CIED-diagnosed AHRE. While this threshold increased the specificity of the AHRE diagnosis, one study reported that almost 20% of the CIED-detected AHRE were not accurate when reviewed by an expert.¹⁷

There are a number of different issues why a CIED can misdiagnose an AHRE episode, which can be classified into false-negative detection (true atrial undersensing because of small EGM signals, functional atrial undersensing because of the EGM signals coinciding with blanking times) and falsepositive detection (myopotential oversensing, electromagnetic interference and lead failure) (Fig. 1).¹²

Fig. 1.

Causes for incorrect AHRE detection by the CIED.

The specificity of the diagnosis also depends on the type of device and the duration of the arrhythmic episode. Therefore, a temporal threshold of five to six minutes was established for several reasons. First of all, to increase the specificity of the diagnosis (decrease the number of false-positive detected episodes).^13,18 Second, episodes longer than five minutes have been shown to increase the risk of stroke.^13,18 Furthermore, certain devices were programmed to only record and classify events longer than a pre-established temporal threshold.

The overall incidence of AHRE in unselected patients is approximately 50%.^13,19-22 However, the studies that excluded patients with a history of AF reported an incidence of approximately 30%.^23-25 The patient population included in most of the above-mentioned studies consisted of elderly patients (mean age > 70 years) with multiple thromboembolic risk factors (mean CHA₂DS₂-VASc score > 2).^13,19-25 An atrial lead is necessary for the CIED in order to accurately diagnose an atrial arrhythmia, which is why single-chamber CIEDs with a ventricular lead have not been included in most studies.

Predictive factors

While AF is still a matter of great interest, the underlying mechanisms that cause and maintain this arrythmia have not been fully understood. Numerous clinical, biological and paraclinical factors have been associated with AF but there are only a handful of studies that examined the role of predictive factors in AHRE.

In the TRENDS study, the incidence of newly detected AHRE did not vary with the CHADS₂ score (CHADS₂ score of 1: 30%, CHADS₂ = 2: 31%, CHADS₂ = 3: 31%) but episodes longer than six hours were associated with an increased CHADS₂ score.²⁴ One study, which included patients with a prior history of AF, showed that older age and increased left atrial volumes were predictors for pacemaker-detected AF.²⁰

An increased percentage of ventricular (VVI) pacing has been associated with an increased risk of developing AF.²⁶ However, even in patients with dual-chamber pacemakers, where atrioventricular synchrony is preserved, an increased percentage of ventricular pacing has been associated with a higher risk of developing AF.^27-35 The most likely explanation is that ventricular pacing causes paradoxical septal motion, which alters interventricular synchrony, lowers ejection fraction and increases filling pressures in the heart chambers. This leads to electric remodelling of the left atrium.

Cumulative ventricular pacing of > 50% has been associated with an increased risk of developing AHRE in patients with no prior history of AF.^23,36 However, one study showed that a high percentage of atrial pacing can also be detrimental. In this study, conducted on patients with no prior history of AF, cumulative atrial pacing > 50% was associated with a three-fold increase in risk of developing AHRE.³⁷

Tekkesin et al. demonstrated that inter-atrial block (IAB) was a predictive factor of AHRE occurrence; 30.1% of the 367 pacemakers implanted for sinus node dysfunction presented AHRE six months after the implantation, at device interrogation. Only 67 patients (27%) in the AHRE-negative group presented with IAB compared to 48 (44.9%) patients in the AHRE-positive group.³⁸ Another study conducted by Rubio Campal et al. also found IAB to be a strong predictor for developing AHRE.³⁹

Although inflammation has been proven to play a certain role in developing and maintaining AF, the underlying mechanism is not fully understood.^40-42,43 Pastori et al. were the first to associate inflammation with an increased risk of developing AHRE.⁴⁴ The results showed that high C-reactive protein and white blood cell count were independently associated with AHRE occurrence. These results suggest a common pathogenetic pathway between AF and AHRE. Another interesting finding of this study was that there was no association between anti-arrhythmic treatment and AHRE incidence, which implies that an optimal level for the management of this arrythmia has not yet been reached.⁴⁴

AF can be a marker of underlying vascular disease because of the direct and indirect mechanisms leading to electrical and anatomical atrial remodelling, which lead to atrial fibrosis.^42,45 The prevalence of CAD in patients with AF ranges from 17 to 46.5%.^46-48 The relationship between vascular disease and AHRE has not been sufficiently investigated. More studies are necessary to investigate the underlying mechanisms and predictive factors of AHRE.

Clinical impact

Atrial high-rate episodes must be distinguished from clinical AF, which is diagnosed by surface ECG and identifies patients with a higher burden of AF. The ancillary MOST analysis was the first study to prove that in CIED patients, AHRE of more than five to six minutes were associated with an increased risk of developing clinical AF.¹⁸ Similar results were reported later by the ASSERT study, where 16% of patients with AHRE longer than six minutes developed clinical AF. These findings suggest a pathogenic connection between these two arrhythmic entities.⁴⁹

In CIED patients, AHRE was associated with a two- to 2.5-fold increase in stroke risk when compared to patients without AHRE.^13,49,50 However, the risk of stroke was smaller than in patients with clinical AF.¹⁴ Kazuo Miyazawa et al. also showed that patients with AHRE had higher mortality rates when compared to patients without AHRE.⁵⁰ However, the West Birmingham Atrial Fibrillation project showed that AHRE did not increase the thromboembolic risk, while the baseline CHA₂DS₂-VASc score was independently associated with thromboembolic events.²⁵

Several burden thresholds have been investigated, ranging from five minutes to 24 hours, to establish a minimum threshold that increases stroke risk. A minimum burden of five minutes has been proven to increase the risk of stroke.^17,18 Pastori et al. showed that AHRE ≥ five minutes were associated with a 1.7-fold increase in major adverse cardiovascular events (MACE) while episodes of more than 24 hours showed a 2.3-fold increase in MACE.⁵¹ However, a re-analysis of the ASSERT study found that only episodes longer than 24 hours were associated with an increased risk of stroke.⁵²

In the SOS AF project, a dichotomised analysis was performed in order to investigate the risk of stroke when comparing different cut-off thresholds (five minutes, one, six, 12 and 23 hours). The one-hour threshold was associated with a hazard ratio of 2.11.53 Data from the RATE registry, a multicentre, prospective, observational study, which investigated outcomes of over 5 000 patients with device-detected AF, showed that short episodes of AF terminating within a single adjudicated EGM recording (up to 20 seconds) did not increase the risk of a composite outcome of stroke, TIA, hospitalisation or mortality. However, approximately 50% of these patients did go on to develop longer episodes of AF over two years of follow up.²²

While the association between AHRE and stroke or systemic thromboembolism has been proven, the minimum duration of an AHRE that increases the thromboembolic risk remains uncertain. Despite the contradictory data, a minimum threshold of five to six minutes is widely considered to increase the risk of thromboembolic events (Table 1).

Table 1. Relationship between CIED-detected AHRE and systemic embolism.

Trial	Number of patients	Follow-up duration	AHRE duration cut-off	Atrial cut-off rate (bpm)	Hazard ratio for TE event (p-value)
Ancillary MOST18	312	27 months (median)	> 5 min	> 220	6.7 (0.020)
Italian AT500 registry59	725	22 months (median)	> 24 h	> 174	3.1 (0.044)
Botto et al.60	568	1 year (mean)	CHADS2 and AF burden . 5 min in a day or > 24 h	> 174	NA
TRENDS13	2486	1.4 years (mean)	≥ 5.5 h	> 175	2.2 (0.060)
Home Monitor CRT61	560	370 days (median)	≥ 3.8 h	> 180	9.4 (0.006)
ASSERT49	2580	2.5 years (mean)	≥ 6 min	> 190	2.5 (0.007)
SOS AF53	10016	2 years (median)	≥ 1 h	> 175	2.11 (0.008)
RATE REGISTRY22	5379	22.9 months (median)	NA	NA	0.87 (0.51)

Despite the temporal relationship between AF and stroke, proven first by the Framingham study and later by the AFFIRM study, there seems to be a temporal discordance between AHRE and stroke.^15,16,54,55 AHRE was diagnosed in only half of the patients with stroke or systemic embolism in an analysis of the TRENDS study.¹⁵ Moreover, 73% of the patients did not present with an AHRE in the preceding month. A similar analysis performed on the ASSERT study identified an AHRE in 51% of the patients with stroke.¹⁶ The cause for this temporal discordance between AHRE and stroke is not completely understood.

AHRE and anticoagulation

Initiation of anticoagulant therapy is difficult in these patients. The threshold of AHRE duration leading to an elevated stroke risk is one of the major knowledge gaps in the EHRA consensus for device-detected subclinical AT.¹² Selecting the most efficient antithrombotic therapy for patients with AHRE was one of the gaps in evidence identified by the European Society of Cardiology (ESC) taskforce for the 2016 guidelines on the management of AF.⁵⁶ There are however ongoing trials that compare oral anticoagulation with aspirin in patients with AHRE9,¹⁰ (Table 2).

Table 2. Comparison of ARTESIA and NOAH trials.

Study	Identifier	Inclusion criteria	Number of patients	Design	Endpoint	Current status	Estimated completion date
ARTESIA9	Clinicaltrials.gov NCT01938248	Patients without clinical AF Pacemaker, ICD or CRT Age ≥ 65 years CHA2DS2-VASc score ≥ 4 1 episode of symptomatic AF ≥ 6 min, atrial rate > 175 bpm no single episode > 24 h in duration	4000	Randomised, doubleblind, double-dummy Randomised to Apixaban 2 × 5 mg or 2 × 2.5 mg vs aspirin 1 × 81 mg	Composite of stroke and SSE Major bleeding	Recruiting	2022
NOAH AFNET 610	Clinicaltrials.gov NCT02618577	Only patients without overt AF Pacemaker or ICD Age ≥ 65 years CHA2DS2-VASc ≥ 2 ≥ 1 episode of AHRE . 6 min, atrial rate > 180 bpm, no single episode > 24 h	2686	Randomised, doubleblind double-dummy Edoxaban 1 × 60 mg or 1 × 30 mg vs 1 × aspirin 100 mg or placebo	Composite of time to first stroke, SSE or CV death	Recruiting	2022

AF, atrial fibrillation; ICD, implantable cardioverter defibrillator; CRT, cardiac resynchronisation therapy; SSE, systemic embolism.

There are no randomised controlled trials published to date to guide anticoagulant therapy in patients with AHRE. All recommendations regarding anticoagulant therapy initiation in these patients are expert recommendations only. Therefore, patient choice regarding this matter is an important consideration.^12,57

The CHA₂DS₂-VASc score seems to also be suitable to assess the stroke risk in patients who present with AHRE.^21,25,58 A recent study, performed on over 21 000 non-anticoagulated patients with CIEDs, showed that the annualised risk of systemic embolism (SSE) was associated with increasing CHA₂DS₂-VASc score and increasing AF duration.

In this study, in patients with a CHA₂DS₂-VASc score of 0 to 1, SSE rates were low, regardless of the duration of the devicedetected AF. However, the stroke risk increased, crossing an actionable threshold, defined as > 1% per year, in patients with CHA₂DS₂-VASc score of 2 with > 23.5 hours of AF, patients with CHA₂DS₂-VASc score of 3–4 and > six minutes of AF, and patients with a CHA₂DS₂-VASc score ≥ 5 even if they presented with no AF.⁵⁸ More studies are necessary to assess if the usual strategies for stroke risk stratification and bleeding risk apply to these patients.

Both the ESC taskforce for the 2016 guidelines on the management of AF and the EHRA consensus for devicedetected subclinical AT recommend using the CHA₂DS₂-VASc score in order to initiate anticoagulation treatment, with similar indications as in AF. Therefore, the ESC taskforce for the 2016 guidelines on the management of AF recommends initiating oral anticoagulation when an AHRE is detected with a duration of more than five to six minutes and an atrial rate of over 180 bpm in male patients with a CHA₂DS₂-VASc score ≥ 1 or in female patients ≥ 2.⁵⁷

The EHRA consensus for device-detected subclinical AT proposes oral anticoagulation based on the same CHA2DS2- VASc values in episodes that are ≥ 5.5 hours.¹² However, the same consensus mentions that an AHRE episode of only minutes has a similar stroke risk as one of > 5.5 hours. The question that arises is whether continuous anticoagulation in these patients is necessary. The 2019 American Heart Association/American College of Cardiology/Heart Rhythm Society (AHA/ACC/ HRS)-focused update of the 2014 AHA/ACC/HRS guideline for the management of patients with AF recommends further evaluation in patients with AHRE to document clinically relevant AF in order to guide the treatment⁵⁶ (Table 3).

Table 3. Society guideline recommendations.

Guideline	Subclinical AF duration	CHA2DS2-VASc score	Class of recommendation
Device-detected subclinical atrial tachyarrhythmias: definition, implications	≥ 5.5 h*	≥ 2	Recommended/indicated
and management; a European Heart Rhythm Association (EHRA) consensus document12	≥ 5.5 h*	1 (men) or 2 (women)	May be used or recommended
ESC 2016 guidelines for the management of atrial fibrillation57	> 5–6 min	≥ 1 in male patients or 2 in female patients	IA

*Data suggests risk is similarly increased by a mere five minutes. AF, atrial fibrillation.

Our approach

In CIED patients we interrogate the device at six weeks after a successful implantation procedure to assess the functioning parameters. Afterwards, we interrogate the devices once a year. If an AHRE episode is observed, we ask for an expert opinion (rhythmology specialist) to confirm that the recorded episode is an AHRE instead of an inaccurately labelled (false-positive) recording. Before considering anti-coagulation therapy, we try to verify the presence of AF by one of the following: resting ECG, Holter ECG recording, patient-operated devices or by reviewing the EGM (if available) to determine if the AHRE was AF. We sometimes use external ECG monitoring devices in CIED patients where the data recorded by the device are uncertain.

After we confirm that the recorded episode is in fact AF, we follow the recommendations of the EHRA consensus for CIED-detected arrythmias and initiate lifelong anticoagulation therapy if the episode was longer than 5.5 hours, based on the patient’s CHA₂DS₂-VASc score (≥ 1 for male patients, ≥ 2 for female patients).¹²

In patients presenting with multiple short episodes of AHRE (≥ five minutes) we follow the same indication, even though the thromboembolic risk is not as high as in the previous group. We therefore initiate anticoagulation therapy based on the CHA₂DS₂-VASc score (≥ 1 for male patients, ≥ 2 for female patients).¹²

In patients with a single short episode of ≥ five minutes, we follow individualised treatment and patient choice based on the thromboembolic risk (CHA₂DS₂-VASc score) and bleeding risk (HASBLED score). We follow the recommendations of the same consensus and observe the AF burden of the patient on multiple follow ups, usually every three to six months, before initiating lifelong anticoagulation. We initiate anticoagulant therapy in this group, only in patients with a high/very high risk of stroke (CHA₂DS₂-VASc score > 4) and low bleeding risk, in which a clear clinical benefit can be anticipated. We do not usually initiate anticoagulant therapy in patients with AHRE < five minutes, because of the lack of data in the literature regarding this duration. Anticoagulant therapy is initiated in these patients on a case-by-case basis.¹² Our strategy always takes into consideration patient choice and wishes.

Conclusion

AHRE represent a complex arrhythmic entity that significantly increases the thromboembolic risk. Further studies are necessary to understand the underlying pathogenic mechanisms behind AHRE and to guide the management of this arrythmia and its complications.