AF burden is important fact or fiction?

Transcription

1 REVIEW ARTICLE AF burden is important fact or fiction? G. Boriani, I. Diemberger, M. Ziacchi, C. Valzania, B. Gardini, P. Cimaglia, C. Martignani, M. Biffi Linked Comment: Lip. Int J Clin Pract 2014; 68: Department of Experimental, Diagnostic and Specialty Medicine, Institute of Cardiology, University of Bologna, S. Orsola-Malpighi University Hospital, Bologna, Italy Correspondence to: Prof. Giuseppe Boriani, MD, PhD, FESC, Department of Experimental, Diagnostic and Specialty Medicine, Institute of Cardiology, University of Bologna, S. Orsola-Malpighi University Hospital, Via Massarenti 9, Bologna, Italy Tel.: Fax: giuseppe.boriani@ unibo.it Disclosures Giuseppe Boriani reports speaker fees from Medtronic and Boston Scientific (small amount); the other authors report no conflicts to disclose. SUMMARY Asymptomatic atrial fibrillation (AF) is common and in view of its prognostic impact (the same as of clinically overt AF) knowledge of the overall AF burden (defined as the amount of time spent in AF) appears to be important, both for scientific and clinical reasons. Data collected on more than 12,000 patients indicate that cardiac implantable electrical devices (CIEDs) are validated tools for measuring AF burden and that AF burden is associated with an increased risk of stroke. A maximum daily AF burden of 1 h carries important negative prognostic implications and may be a clinically relevant parameter for improving risk stratification for stroke. Decision-making should primarily consider the context in which asymptomatic, subclinical arrhythmias are detected (i.e. primary or secondary prevention of stroke and systemic embolism) and the risk profile of every individual patient with regard to thromboembolic and haemorrhagic risk, as well as patient preferences and values. Continuous monitoring using CIEDs with extensive data storage capabilities allow in-depth study of the temporal relationship between AF and ischaemic stroke. The relationships between AF and stroke are complex. AF is certainly a risk factor for cardioembolic stroke, with a cause effect relationship between the arrhythmia and a thromboembolic event, the latter being related to atrial thrombi. However, AF can also be a simple marker of risk, with a non-causal association between the arrhythmia and stroke, the latter being possibly related to atheroemboli from the aorta, the carotid arteries or from other sources. Introduction Atrial fibrillation (AF) is the most common sustained arrhythmia encountered in clinical practice and is associated with significant morbidity and adverse outcomes (stroke, heart failure, death). The clinical approach to AF should consider the wide range of clinical presentations of this arrhythmia. As a matter of fact, in some patients AF is apparently the disease itself (e.g. recurrent paroxysmal lone AF in a healthy patient) or is associated with minor disorders (high normal thyroid function, subclinical atherosclerosis), facilitating factors (alcohol abuse, smoking, drugs, etc.), or other conditions facilitating its occurrence (hypertension, diabetes, obesity, sleep apnea, etc.) (1,2). In most patients, AF is associated with underlying structural heart disease (coronary heart disease, valvular diseases, dilated cardiomyopathy or other myocardial dysfunctions leading to heart failure, etc.) or occurs in a patient with a major extracardiac disease (advanced chronic kidney disease, neoplasm, etc.) (1 4). However, in any clinical scenario, the starting point of management is the detection of AF and Review criteria We searched MEDLINE and Google Scholar using the terms atrial fibrillation burden and stroke or thromboembolism and manually reviewed the references in the English language. Abstracts from meetings in the field of cardiology were analysed in order to find unpublished data. Message for the clinic Knowledge of the overall arrhythmic burden of AF (defined as the amount of time spent in AF, either symptomatic or not) appears to be important for scientific and clinical reasons. Cardiac implantable electrical devices are validated tools for measuring AF burden, which is associated with an increased risk of stroke. Decision-making should primarily consider the clinical context and the risk profile with regard to thromboembolic and haemorrhagic risk. confirmation of the arrhythmia through an electrocardiographic recording, usually a 12-lead electrocardiogram (ECG) (4). This step is the basis for assessing the thromboembolic risk and making decisions on the basis of risk stratification for stroke, also considering the haemorrhagic risk of antithrombotic agents (4 6). This is a cornerstone of AF management that should be followed by a series of clinical considerations on the pattern of AF and associated symptoms should suggest the choice between a ratecontrol or a rhythm-control strategy (2,4,7). A growing number of scientific articles is dedicated to the various aspects of AF epidemiology and management and the term AF burden has been repeatedly used, in different contexts and with different meanings, as reported in Table 1. AF burden as a measure of arrhythmia presence and duration: insights into the complex issue of silent AF The term AF burden has recently been applied to describe the temporal dynamic pattern of AF, in terms of presence and duration of AF episodes, as 444. doi: /ijcp.12326

2 Atrial fibrillation burden and stroke 445 Table 1 Application of the concept of burden to different aspects of AF, according to the literature Epidemiological burden Global burden Financial burden Economic burden Hospitalisation burden Caregiver burden Disease burden Clinical burden Arrhythmia burden detected by continuous monitoring with an implanted device. Via an atrial lead, cardiac implantable electrical devices (CIEDs) are capable of monitoring the atrial rhythm and storing the data on atrial tachyarrhythmias and AF episodes, thus providing detailed information on the arrhythmia presence, duration of every episode, time of occurrence and overall time spent in AF during a specified period of time. In this context, a landmark study was published by Israel et al. (8) who clearly outlined how, in a selected population of 110 patients with previous paroxysmal or persistent AF implanted with a pacemaker with extended diagnostic capabilities for AF (Medtronic AT 500, Medtronic Inc., Minneapolis, MN, USA), AF episodes of considerable duration (> 48 h) can be diagnosed by the device much more frequently than with conventional regular ECG follow up (in 88% vs. 46% of patients). Moreover, the same study highlighted that episodes of AF lasting more than 48 h may be completely asymptomatic and may occur even after several weeks or months of freedom from AF. This study and a series of subsequent reports clearly showed that a considerable number of AF episodes is asymptomatic, or clinically silent and that patients may experience both symptomatic and asymptomatic AF episodes of variable duration, with a ratio of at least 12 asymptomatic per 1 symptomatic episode, according to some Holter studies (9). The exact number of patients with silent AF is unknown, but it has been estimated that among patients with recognised AF, one-third has no appreciable symptoms. Since silent asymptomatic AF carries the same prognostic significance as symptomatic AF in terms of hard outcomes such as stroke and death (10), population screening has been proposed with various methods ranging from simple pulse palpation, single-lead ECG, Holter recordings of variable duration, up to newer technologies, applied to smartphones (11,12). When detection of AF is a clinical priority, such as in patients with a previous stroke of suspected cardioembolic origin, but without documentation of paroxysmal AF, the use of specific ECG monitoring strategies is justified, but methods of intermittent monitoring (with conventional Holter recordings of h, longer Holter recordings, up to 30 days or with transtelephonic transmissions of the ECG) have limited sensitivity compared with continuous monitoring using an implanted device (13) (Figure 1). Cardiac implantable electrical devices with an atrial lead make it possible to detect atrial high-rate episodes (AHRE), corresponding to any atrial tachyarrrhythmia above a predefined rate threshold (> bpm), thus including AF, atrial flutter and regular atrial tachycardias (14). Detection of AHRE is related to a number of technical issues, most of which are manufacturer-specific, as well as to the atrial sensitivity and device programming with regard to atrial rate and episode duration cut-offs. Although temporal cut-offs for detection and storage of AHRE data as short as s have been used, the diagnostic accuracy is highly reliable when episodes > 5 min in duration are considered, thus obtaining appropriate detection of 95% of AF episodes and minimising the risk of detecting artefacts caused by myopotentials or other sources of electrical interference (15,16). In a single patient presenting with one or more episodes of silent AF at a routine device check, the possibility to review the electrograms (EGMs) of the tachyarrhythmia stored in the device s memory makes it possible to confirm the device diagnosis of AF and to rule out potential causes of oversensing in the atrial channel (2,11), thus reaching a specific confirmation of AF diagnosis. Figure 1 Evaluation in simulated conditions of the sensitivity of intermittent ECG and Holter recordings vs. continuous monitoring through an implanted dual chamber pacemaker. The sensitivity of intermittent recordings in detecting AF episodes lasting > 24 h is very low, below 40%. Modified from Botto et al. (13), with permission

3 446 Atrial fibrillation burden and stroke As a consequence of findings from the studies focused on atrial tachyarrhythmias and AHRE, the concept of atrial burden has been introduced, with a number of different technical definitions, as shown in Table 2. More recently, AF burden has been defined as the amount of time spent in AF per day of a specified follow-up period ( daily AF burden ) and the maximum daily AF burden (the highest daily burden observed in a long follow-up period) has been the subject of several studies aimed at assessing the temporal distribution of AF, its progression and response to antiarrhythmic interventions as well as the association with an increased risk of thromboembolic events and stroke. AF burden detected by CIEDs: a way of studying the temporal distribution of AF, its progression and response to antiarrhythmic interventions? The temporal patterns of AF burden progression and the conversion of paroxysmal to persistent AF were evaluated by Saksena et al. (17) in a group of patients with a history of paroxysmal AF, implanted with a pacemaker for bradyarrhythmias. Progression to persistent AF was observed in one-fourth of cases, after a mean period of approximately 4 months, with a progressive increase in AF burden, specifically in patients with structural heart disease. These findings stress that AF is a progressive disease and this should have important implications when evaluating the study design of trials focused on rhythm control strategies, whereby a crossover design should be avoided. Moreover, since in more than 60% of patients AF recurrences do not follow a uniform distribution, the conventional use of time to first AF recurrence as an end-point of efficacy appears questionable (18). Other studies, also based on modelling, confirm the limitations of trials evaluating the response to rhythm-control treatments in terms of AF burden reduction since temporal fluctuations in AF burden can be so high that about 60% of patients could falsely appear as responders or non-responders in a crossover study, regardless of AF burden reduction (19,20). Monitoring of AF episodes and AF burden can be of great importance also for patients treated with AF ablation, since the assessment of asymptomatic recurrences may better represent the patient s response and become a crucial step for taking into consideration the possibility of discontinuation of anticoagulation, in case of complete cure of AF, a strategy which is still under investigation and not advised by current consensus guidelines (4,5). Anyway, the interest in the use of continuous monitoring of AF recurrences after left atrial ablation, as a more reliable tool as compared with conventional Holter recordings or long-duration Holter, is supported by the evidence that the proportion of asymptomatic episodes increase after ablation (21). Martinek et al. reported on a group of 14 patients implanted with a pacemaker with extended diagnostic capabilities for AF (Medtronic AT500) and followed after AF ablation (22). A proportion of patients ranging from 57% to 71% was considered as responding to ablation on the basis of reduction in symptomatic episodes, or intermittent Holter recordings (from 24 h to 7-day) but the proportion of responders dropped to 43%, considering a significant AF reduction, or to 21% considering complete long-term atrial tachyarrhythmia suppression, as documented by the implanted device (22). Pokushalov et al. systematically used implantable cardiac monitors (ICM) to follow patients with AF treated with left atrial ablation and found that an AF burden of > 4.5% detected during the 2 months postablation blanking period was a powerful predictor of ablation failure (23). According to these authors, continuous AF monitoring can be proposed for guiding postablation management, since it may guide antiarrhythmic therapy, indication to re-intervention and, potentially, antithrombotic treatment. Dual chamber pacemakers have recently been used to study the efficacy of antiarrhythmic drugs in controlled studies that measured AF burden, in patients previously implanted with a DDD pacemaker (24,25). The systematic use of such an approach for the evaluation of drug treatments is under evaluation by regulatory agencies. Table 2 Technical definitions of AF burden historically used in trials on CIEDs (pacemakers or ICDs) Percentage of time spent in AF during a follow-up period (% time in AF) Daily AF burden as the time spent in AF during a single day of the follow-up period (from this measurement, the maximum daily burden during a period of observation can be assessed) Average number of hours per day spent in AF Cumulative duration of AF recurrences during a follow-up period (e.g. 6 months) Total number of days with AF divided by the cumulative follow-up days

4 Atrial fibrillation burden and stroke 447 AF burden detected by CIEDs: a way of improving prevention of thromboembolic events and stroke? The more accurate and extended diagnostic capabilities of CIEDs have highlighted the occurrence of silent AF in many patients treated with a device (a pacemaker or an ICD) but the key question in the clinical setting is what amount of AF, or what threshold of AF burden, is associated with a significant risk of stroke which justifies the initiation of an antithromboembolic prophylaxis (with warfarin or a novel anticoagulant) according to the widely accepted risk stratification scores, such as the CHADS 2 and CHA 2 DS 2 -VASc scores? An overview of the studies reporting on the association of different AF burden thresholds with stroke (13,26 33), which can help to find an answer to this challenging question, is shown in Table 3. The first study reporting on the prognostic implications of the diagnostic features of implantable devices was a sub-analysis of the MOde Selection Trial (MOST) (26). In this sub-study, which evaluated 312 patients, the presence of AHRE > 5 min in duration was associated with a significantly increased risk of death with a hazard ratio (HR) of 2.48, or non-fatal stroke with a HR of For the Italian AT500 Registry, Capucci et al. reported data from 725 patients in whom devicedetected AT/AF episodes > 24 h in duration were associated with a significantly increased risk of stroke or systemic embolism with a HR of 3.1 (27). In an additional study from the same study group, Botto et al. showed that risk stratification for thromboembolic events can be improved by combining CHADS 2 score with device-detected data on AF presence/duration, on the basis of a lower threshold for AF of 5 min (13). Additional data were provided by TRENDS, a prospective, observational study enrolling patients with > 1 stroke risk factor (heart failure, hypertension, age 65, diabetes, or prior thromboembolism) receiving pacemakers or defibrillators able to measure AF burden (defined as the longest total AT/AF Table 3 Studies in the literature that analysed the relationship between AHRE or AF burden, as detected by an implanted CIEDs (a pacemaker or an ICD) Author, year, reference No. of patients AF burden associated with stroke HR (95% CI) for stroke p-value Other findings Glotzer et al., 2003 (26) 312 (patients with sinus node dysfunction) 5 min 2.79 ( ) p = Capucci et al., 2005 (27) 725 (patients with bradyarrhythmias and history of PAF) > 24 h 3.1 ( ) p = Botto et al., 2008 (13) 568 (patients with bradyarrhythmias and history of PAF) > 5 min Combining AF burden and CHADS 2 make it possible to distinguish a subgroup at low and high risk of stroke Glotzer et al., 2009 (28) 2486 (patients with 1 stroke risk factor implanted with a pacemaker or an ICD) 5.5 h 2.20 ( ) p = 0.06 Ziegler et al., 2010 (29) 163 (previous thromboembolic event, no PAF) 5 min 73% of new AF patients with previous TE experienced episodes of AF < 10% of follow-up days Boriani et al., 2011 (30) Healey et al., 2012 (31) Shanmugam et al., 2012 (32) Boriani et al (33) 568 (patients with bradyarrhythmias and history of PAF) 2580 ( 65 years, hypertension, no history of PAF) 560 (heart failure patients treated with CRT) 3.8 h 9.4 ( ) 10,016 patients with a CIED, without permanent AF, median age 70 years (pooled analysis of three studies) > 5 min Combining AF burden and CHADS 2 or CHA 2 DS 2 -VASc improves prediction of stroke, reaching C-statistics of and 0.910, respectively > 6 min 2.49 ( ) p = % of the study population had at least p = day with AF burden > 14 min 1 h 2.11 ( ) p = PAF, paroxysmal atrial fibrillation; CIED, cardiac implantable electrical device; HR, hazard ratio; CI, confidence intervals.

5 448 Atrial fibrillation burden and stroke duration on any given day during the prior 30-day period) (28). In this study, if the AF burden was 5.5 h, the adjusted HR for thromboembolism was 2.20 ( , p = 0.06) (28). Ziegler et al. (29) showed that newly detected episodes of AF were found via continuous monitoring in 28% of patients with previous thromboembolism; however because of temporal distribution most episodes would not have been detected by standard intermittent monitoring techniques. Boriani et al. (30) reported how implementation of device data on AF presence/duration/burden has the potential to contribute to improved clinical risk stratification. The addition of AF burden improved c-statistics: for CHADS 2 from (p = 0.051) to (p = 0.007); for CHA 2 DS 2 -VASc, from (p < ) to (p < ). The ASSERT study, reported by Healey et al. (31), enrolled 2580 patients, 65 years of age or older, with hypertension and no history of AF, implanted with a pacemaker or an ICD. During a 3-month period, subclinical atrial tachyarrhythmias (atrial rate > 190 bpm for > 6 min) occurred in around 10% of patients and were associated with an increased risk of clinical and of ischaemic stroke or systemic embolism (HR, 2.49) during a 2.5-year follow up. As stressed in the accompanying editorial (34), the important data from ASSERT do not clarify what threshold of AF burden may justify treatment with warfarin or novel anticoagulants. Shanmugam et al. reported on 360 heart failure patients implanted with a device for cardiac resynchroniation therapy (CRT) with remote monitoring capabilities (32). A device-detected atrial burden 3.8 h in a day (using a cut-off of 180 bpm) was associated with an increased incidence of thromboembolic events as well as with an increased occurrence of the composite end-point of admissions for AF, heart failure, thromboembolism or cardiovascular death. Recently, a pooled analysis of individual patient data from three prospective studies was performed in the SOS AF project (33), collecting an overall population of 10,016 patients (median age 70 years). During a median follow up of 24 months, 43% of patients experienced at least 1 day with at least 5 min of AF burden. A Cox regression analysis adjusted for CHADS 2 score and anticoagulants at baseline demonstrated that AF burden was an independent predictor of stroke and a threshold of 1 h was associated with the highest HR for stroke, i.e (95% CI , p = 0.008) in a dichotomised analysis that compared other potential threshold cut-offs for AF burden (5 min, 1, 6, 12 and 23 h, respectively). It is not known what is the propensity of clinicians to take appropriate clinical decisions on the basis of device data on AF burden. The ANGELS of AF project was a medical care program aimed at supporting adherence to oral anticoagulation guidelines for thromboprophylaxis through the use of ICD AF diagnostics (35). Fifty Italian cardiology clinics followed 3438 patients with ICDs. In a subgroup of 15 centres (the ANGELS of AF centres), cardiologists attending follow-up visits were supplied with specific reports describing stroke risk factors and risk scores (CHADS 2 ). Other centres represented a control group of patients as a comparison of oral anticoagulants use. In the ANGELS of AF centres, the percentage of patients on oral anticoagulant therapy, as indicated by guidelines, increased during follow up from around 46% at baseline, to up to around 73% at the end of the observation period, with many more patients on anticoagulants as compared with control patients. The perspective of using continuous monitoring through CIEDs not only for initiating oral anticoagulants, in patients at risk, but also for discontinuing it, as a result of apparently complete efficacy of rhythm control therapies (AF ablation or antiarrhythmic drugs) resulting in so called AF cure, has been hypothesised, but at present is still of debatable safety and is under investigation (36,37). AF burden and the relationship between AF and stroke: association or causation? Continuous monitoring through a CIED with the ability to store data on all the episodes of sustained AF allows an in-depth study of the temporal relationship between AF and ischaemic stroke. A series of studies focused on patients with AF implanted with a CIED found that ischaemic stroke may occur without concurrent presence of atrial tachyarrhythmias or AF at the time of stroke or in the days before. In detail, the study by Daoud et al. (38) showed that 45% of the patients with a devicedetected atrial tachyarrhythmia prior to the ischaemic event (stroke or cerebrovascular embolism) did not have any arrhythmia in the 30 days before the event. Similar findings were reported by the ANGELS of AF study, where among 33 patients with stroke, transient ischaemic attacks, or embolic events, 64% had an AF burden > 5 min, detected by ICD diagnostics, at any time prior to stroke, 33% in the 30 days preceding the event (35). According to these data, the relationship between AF and stroke need to be reconsidered, in all its complexity. AF is certainly a risk factor for cardioembolic

6 Atrial fibrillation burden and stroke 449 stroke, with a cause effect relationship between the arrhythmia and a thromboembolic event, the latter being related to atrial thrombi (14). However, on the basis of a series of clinical observations, AF can also be a simple marker of risk, with a non-causal association between the arrhythmia and stroke, the latter being possibly related to atheroemboli from the aorta, the carotid arteries or from other sources (39 42). Perspectives: continuous monitoring of cardiac rhythm through remote monitoring of CIEDs The conventional in-office check of implanted pacemakers, ICD or CRT devices is a time-consuming procedure (43). The evolution of electronic technology of telemedicine now allows remote monitoring of patients implanted with CIEDs, also in the perspective of a shift from device-centred to patientoriented telemonitoring, and disease management (44,45). The implementation of remote monitoring requires new models of health-care organisation, with interesting economic perspectives (46,47). Many multicentre studies showed how remote monitoring of CIEDs makes it possible to significantly shorten the time to clinical decisions with regard to information on device and patient status that also includes presence and duration of AF burden (45,48 50). However, it is noteworthy that, despite the important and potentially cost-saving perspectives, remote monitoring is not included in reimbursed activities, or is not adequately reimbursed, in most European countries, thus creating an important barrier to widespread implementation in daily practice (51). Detection of previously unrecognised clinically silent AF is a typical condition where appropriate decision-making may have important implications for patient outcome. This is strongly supported by the evidence that the risk of stroke is the same both in asymptomatic and symptomatic AF (10). In a modelling study, Ricci et al. (52) calculated that prompt reaction to AF detection with appropriate, evidence-based decision-making on antithrombotic treatment may result in an important reduction of strokes, in the range of 9 18% at 2 years. Therefore, it is possible to hypothesise that using diagnostic data remotely transmitted from CIEDs, for improving antithrombotic prophylaxis, may represent a win-win perspective where many stakeholders (patients, physicians, healthcare providers, device manufacturers) can obtain a substantial benefit (53). Figure 2 shows an example of remote transmission of data on AF burden from a patient (implanted with a pacemaker for bradyarrhythmias). Specific features such as the alarms make it possible to customise data notification as well as to obtain a clear picture of new detected AF occurring in a single patient or in a group of patients followed by a dedicated clinic. The benefit of early detection of AF or of a significant AF burden can be maximised in patients implanted with a device for CRT, a context where AF not only increases the thromboembolic risk but is also associated with loss of biventricular pacing, and therefore with loss of the favourable clinical and economic benefits (45,49,54). Perspectives: continuous monitoring of cardiac rhythm through subcutaneous ICMs Continuous monitoring through an implanted device with an atrial lead makes it possible to obtain a sensitivity in AF diagnosis superior to any method based on intermittent monitoring, even including longterm Holter recordings (13,55). In order to provide continuous monitoring of the cardiac rhythm with a leadless system, subcutaneous ICMs have been developed and tested (11). In these devices, the diagnosis of AF is based on the analysis of consecutive RR intervals and irregularity of the RR interval is assessed in a Lorentz plot, where each RR interval is plotted against the previous value of the RR interval (11,21). The diagnostic performance of the Reveal XT (Medtronic Inc., Minneapolis, MN, USA) ICM was assessed in a dedicated study and the result was an accurate measure of AF burden (in 98.5% of patients with AF diagnosed with conventional Holter) and a sensitivity and specificity for identifying a patient with AF of 96.1% and 85.4%, respectively (21). Atrial fibrillation is found incidentally in about 25% of admissions for stroke. Since detection of AF in patients with cardioembolic stroke makes it possible to institute oral anticoagulants (56,57), with marked benefit in terms of prevention of stroke recurrence, intermittent and, more recently, continuous monitoring with use of implantable loop recorders or ICMs has been proposed and is under investigation (11,55,58). A series of controlled trials is ongoing in patients presenting with cryptogenic stroke and there is preliminary evidence that in selected patients AF can be detected by an ICM in % of cryptogenic or unexplained strokes (59 61). It is expected that ongoing and future studies will define in what subset of patients presenting with stroke implant of an ICM for detecting AF and measuring AF burden is clinically useful and cost-effective.

7 450 Atrial fibrillation burden and stroke Figure 2 Data obtained through remote monitoring though CareLink network (Medtronic) in a patient previously implanted with a pacemaker for bradyarrhythmias. In the top panel, diagnostic data indicate the temporal distribution of episodes of atrial tachyarrhythmias (AT/AF), as well as% of atrial and ventricular pacing. The report indicates that since the last follow up, the patient had 3 days with more than 6 h of AF (i.e. the patient had a daily burden of more than 6 h). In the bottom panel, specific data on time of occurrence and duration of every detected episode of atrial tachyarrhythmia are reported, together with possibility to analyse recorded electrograms (EGMs) of the most recent episodes. In these patients, AF episodes lasted from few minutes to several hours and were asymptomatic Conclusions Asymptomatic AF is common and in view of its prognostic impact, which is the same as of clinically overt AF, knowledge of the overall arrhythmic burden of AF appears to be important, both for scientific and clinical reasons. According to the amount of data collected by many studies in recent years, detailed infor-

8 Atrial fibrillation burden and stroke 451 mation on AF burden should now be considered as a fact, not as a fiction. Data collected on more than 12,000 patients indicate that implantable devices are validated tools for measuring AF burden and that AF burden, as detected using the diagnostic features of CI- EDs, is associated with an increased risk of stroke. The specific threshold of AF burden associated with a substantial increase in the risk of stroke is still a matter of investigation and debate, but it is clear that a maximum daily AF burden of 1 h carries important negative prognostic implications and may be a clinically relevant parameter for improving risk stratification for stroke. In our view, when data on device-detected AF (in terms of AF burden) become available, at a routine device check or through remote monitoring, the approach should be mainly clinical, based on a complete assessment of the clinical status, coupled with revision of stored EGM of the arrhythmia. Decisionmaking should primarily consider the context where subclinical arrhythmias are detected (i.e. primary or secondary prevention of stroke and systemic embolism) and the risk profile of every individual patient, with regard to thromboembolic and haemorrhagic risk. In other words, data on the presence/absence of AF and on its quantification in terms of AF burden (minutes to hours of daily burden) should prompt an individualised and tailored approach, where available scientific evidence is combined with device data, also taking into account patient preferences and values. Author contributions Giuseppe Boriani: concept/design, data interpretation, drafting article. Igor Diemberger: literature search, drafting sections of the article, critical revision of article, approval of article. Matteo Ziacchi: literature search, drafting sections of the article, critical revision of article, approval of article. Cinzia Valzania: literature search, critical revision of article, approval of article. Beatrice Gardini: data interpretation, drafting sections of the article, critical revision of article, approval of article. Paolo Cimaglia: literature search, critical revision of article, approval of article. Cristian Martignani: data interpretation, drafting sections of the article, critical revision of article, approval of article. Mauro Biffi: data interpretation, drafting sections of the article, critical revision of article, approval of article. References 1 Boriani G, Diemberger I, Martignani C, Biffi M, Branzi A. The epidemiological burden of atrial fibrillation: a challenge for clinicians and health care systems. Eur Heart J 2006; 27(8): Kirchhof P, Lip GY, Van Gelder IC et al. Comprehensive risk reduction in patients with atrial fibrillation: emerging diagnostic and therapeutic options a report from the 3rd Atrial Fibrillation Competence NETwork/European Heart Rhythm Association consensus conference. Europace 2012; 14(1): Boriani G, Cervi E, Diemberger I, Martignani C, Biffi M. Clinical management of atrial fibrillation: need for a comprehensive patient-centered approach. Chest 2011; 140(4): Camm AJ, Kirchhof P, Lip GY et al. Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Eur Heart J 2010; 31(19): Camm AJ, Lip GY, De Caterina R et al.; ESC Committee for Practice Guidelines (CPG) focused update of the ESC Guidelines for the management of atrial fibrillation: an update of the 2010 ESC Guidelines for the management of atrial fibrillation. Developed with the special contribution of the European Heart Rhythm Association. Eur Heart J 2012; 33(21): Lip GY. Stroke and bleeding risk assessment in atrial fibrillation: when, how, and why? Eur Heart J 2013; 34(14): Boriani G, Biffi M, Diemberger I, Martignani C, Branzi A. Rate control in atrial fibrillation: choice of treatment and assessment of efficacy. Drugs 2003; 63(14): Israel CW, Gr onefeld G, Ehrlich JR, Li YG, Hohnloser SH. Long-term risk of recurrent atrial fibrillation as documented by an implantable monitoring device: implications for optimal patient care. J Am Coll Cardiol 2004; 43(1): Savelieva I, Camm AJ. Clinical relevance of silent atrial fibrillation: prevalence, prognosis, quality of life, and management. J Interv Card Electrophysiol 2000; 4(2): Flaker GC, Belew K, Beckman K et al. Asymptomatic atrial fibrillation: demographic features and prognostic information from the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) study. Am Heart J 2005; 149(4): Camm AJ, Corbucci G, Padeletti L. Usefulness of continuous electrocardiographic monitoring for atrial fibrillation. Am J Cardiol 2012; 110(2): McManus DD, Lee J, Maitas O et al. A novel application for the detection of an irregular pulse using an iphone 4S in patients with atrial fibrillation. Heart Rhythm 2013; 10(3): Botto GL, Padeletti L, Santini M et al. Presence and duration of atrial fibrillation detected by continuous monitoring: crucial implications for the risk of thromboembolic events. J Cardiovasc Electrophysiol 2009; 20(3): Khoo CW, Krishnamoorthy S, Lim HS, Lip GY. Atrial fibrillation, arrhythmia burden and thrombogenesis. Int J Cardiol 2012; 157(3): Pollak WM, Simmons JD, Interian A Jr et al. Clinical utility of intraatrial pacemaker stored electrograms to diagnose atrial fibrillation and flutter. Pacing Clin Electrophysiol 2001; 24(4 Pt 1): Purerfellner H, Gillis AM, Holbrook R, Hettrick DA. Accuracy of atrial tachyarrhythmia detection in implantable devices with arrhythmia therapies. Pacing Clin Electrophysiol 2004; 27(7): Saksena S, Hettrick DA, Koehler JL, Grammatico A, Padeletti L. Progression of paroxysmal atrial fibrillation to persistent atrial fibrillation in patients with bradyarrhythmias. Am Heart J 2007; 154(5): Ricci R, Santini M, Padeletti L et al. Atrial tachyarrhythmia recurrence temporal patterns in bradycardia patients implanted with antitachycardia pacemakers. J Cardiovasc Electrophysiol 2004; 15(1): Padeletti L, Santini M, Boriani G et al. Temporal variability of atrial tachyarrhythmia burden in bradycardia-tachycardia syndrome patients. Eur Heart J 2005; 26(2): Botto GL, Santini M, Padeletti L et al. Temporal variability of atrial fibrillation in pacemaker recipients for bradycardia: implications for crossover designed trials, study sample size, and identification of responder patients by means of arrhythmia burden. J Cardiovasc Electrophysiol 2007; 18(3): Hindricks G, Piorkowski C, Tanner H et al. Perception of atrial fibrillation before and after radiofrequency catheter ablation: relevance of asymptomatic arrhythmia recurrence. Circulation 2005; 112(3): Martinek M, Nesser HJ, Aichinger J, Boehm G, Purerfellner H. Accuracy of integration of multislice computed tomography imaging into three-dimensional electroanatomic mapping for real-time guided radiofrequency ablation of left atrial fibrillation-influence of heart rhythm and radiofrequency lesions. J Interv Card Electrophysiol 2006; 17(2):

9 452 Atrial fibrillation burden and stroke 23 Pokushalov E, Romanov A, Corbucci G et al. Does atrial fibrillation burden measured by continuous monitoring during the blanking period predict the response to ablation at 12-month follow-up? Heart Rhythm 2012; 9(9): Arya A, Silberbauer J, Teichman SL, Milner P, Sulke N, Camm AJ. A preliminary assessment of the effects of ATI-2042 in subjects with paroxysmal atrial fibrillation using implanted pacemaker methodology. Europace 2009; 11(4): Ezekowitz MD, Nagarakanti R, Lubinski A et al. A randomized trial of budiodarone in paroxysmal atrial fibrillation. J Interv Card Electrophysiol 2012; 34(1): Glotzer TV, Hellkamp AS, Zimmerman J et al.; MOST Investigators. Atrial high rate episodes detected by pacemaker diagnostics predict death and stroke: report of the Atrial Diagnostics Ancillary Study of the MOde Selection Trial (MOST). Circulation 2003; 107(12): Capucci A, Santini M, Padeletti L et al.; Italian AT500 Registry Investigators. Monitored atrial fibrillation duration predicts arterial embolic events in patients suffering from bradycardia and atrial fibrillation implanted with antitachycardia pacemakers. J Am Coll Cardiol 2005; 46(10): Glotzer TV, Daoud EG, Wyse DG et al. The relationship between daily atrial tachyarrhythmia burden from implantable device diagnostics and stroke risk: the TRENDS study. Circ Arrhythm Electrophysiol 2009; 2(5): Ziegler PD, Glotzer TV, Daoud EG et al. Incidence of newly detected atrial arrhythmias via implantable devices in patients with a history of thromboembolic events. Stroke 2010; 41(2): Boriani G, Botto GL, Padeletti L et al.; Italian AT-500 Registry Investigators. Improving stroke risk stratification using the CHADS 2 and CHA 2 DS 2 -VASc risk scores in patients with paroxysmal atrial fibrillation by continuous arrhythmia burden monitoring. Stroke 2011; 42(6): Healey JS, Connolly SJ, Gold MR et al. Subclinical atrial fibrillation and the risk of stroke. N Engl J Med 2012; 366(2): Shanmugam N, Boerdlein A, Proff J et al. Detection of atrial high-rate events by continuous home monitoring: clinical significance in the heart failure-cardiac resynchronization therapy population. Europace 2012; 14(2): Boriani G, Glotzer TV, Santini M et al. Device detected atrial filbrillation and risk for stroke: an analysis of more than 10,000 patients from the SOS AF project (Stroke Prevention Strategies based on Atrial Filbrillation inform from implanted devices). Eur Heart J 2013;. doi: /eurheartj/eht Lamas G. How much atrial fibrillation is too much atrial fibrillation? N Engl J Med 2012; 366(2): Boriani G, Santini M, Lunati M et al. Improving thromboprophylaxis using atrial fibrillation diagnostic capabilities in implantable cardioverter-defibrillators: the multicentre Italian ANGELS of AF Project. Circ Cardiovasc Qual Outcomes 2012; 5: Lim HS, Lip GY. Asymptomatic atrial fibrillation on device interrogation. J Cardiovasc Electrophysiol 2008; 19(8): Ip J, Waldo AL, Lip GY et al.; IMPACT Investigators. Multicenter randomized study of anticoagulation guided by remote rhythm monitoring in patients with implantable cardioverter-defibrillator and CRT-D devices: rationale, design, and clinical characteristics of the initially enrolled cohort The IMPACT study. Am Heart J 2009; 158(3): Daoud EG, Glotzer TV, Wyse DG et al.; TRENDS Investigators. Temporal relationship of atrial tachyarrhythmias, cerebrovascular events, and systemic emboli based on stored device data: a subgroup analysis of TRENDS. Heart Rhythm 2011; 8(9): The French Study of Aortic Plaques in Stroke Group. Atherosclerotic disease of the aortic arch as a risk factor for recurrent ischemic stroke. N Engl J Med 1996; 334: Benjamin EJ, Plehn JF, D Agostino RB et al. Mitral annular calcification and the risk of stroke in an elderly cohort. N Engl J Med 1992; 327(6): Stein JH, Soble JS. Thrombus associated with mitral valve calcification: a possible mechanism for embolic stroke. Stroke 1995; 26: Chugh SS, Blackshear JL, Shen WK, Hammill SC, Gersh BJ. Epidemiology and natural history of atrial fibrillation: clinical implications. J Am Coll Cardiol 2001; 37(2): Boriani G, Auricchio A, Klersy C et al.; European Heart Rhythm Association; Eucomed. Healthcare personnel resource burden related to in-clinic follow-up of cardiovascular implantable electronic devices: a European Heart Rhythm Association and Eucomed joint survey. Europace 2011; 13(8): Boriani G, Diemberger I, Martignani C et al. Telecardiology and remote monitoring of implanted electrical devices: the potential for fresh clinical care perspectives. J Gen Intern Med 2008; 23(Suppl. 1): Varma N, Ricci RP. Telemedicine and cardiac implants: what is the benefit? Eur Heart J 2013; 34 (25): Burri H, Heidb uchel H, Jung W, Brugada P. Remote monitoring: a cost or an investment? Europace 2011; 13(Suppl. 2): ii Boriani G, Maniadakis N, Auricchio A et al. Health technology assessment in interventional electrophysiology and device therapy: a position paper of the European Heart Rhythm Association. Eur Heart J 2013; 34(25): Crossley GH, Boyle A, Vitense H, Chang Y, Mead RH; CONNECT Investigators. The CONNECT (Clinical Evaluation of Remote Notification to Reduce Time to Clinical Decision) trial: the value of wireless remote monitoring with automatic clinician alerts. J Am Coll Cardiol 2011; 57(10): Landolina M, Perego GB, Lunati M et al. Remote monitoring reduces healthcare use and improves quality of care in heart failure patients with implantable defibrillators: the evolution of management strategies of heart failure patients with implantable defibrillators (EVOLVO) study. Circulation 2012; 125(24): Boriani G, Da Costa A, Ricci RP et al. The MOnitoring Resynchronization devices and CARdiac patients (MORE-CARE) randomized controlled trial: phase 1 results on early decision making with remote monitoring. J Med Internet Res 2013; 15(8): e Boriani G, Burri H, Mantovani LG et al. Device therapy and hospital reimbursement practices across European countries: a heterogeneous scenario. Europace 2011; 13(Suppl. 2): ii Ricci RP, Morichelli L, Gargaro A, Laudadio MT, Santini M. Home monitoring in patients with implantable cardiac devices: is there a potential reduction of stroke risk? Results from a computer model tested through Monte Carlo simulations. J Cardiovasc Electrophysiol 2009; 20(11): Boriani G, Diemberger I, Mantovani V, Biffi M, Martignani C. Remote monitoring of patients with an implanted device and patients outcomes: the potential for win-win dynamics. J Cardiovasc Electrophysiol 2009; 20(11): Boriani G, Diemberger I, Biffi M, Martignani C. Cost-effectiveness of cardiac resynchronisation therapy. Heart 2012; 98(24): Seet RC, Friedman PA, Rabinstein AA. Prolonged rhythm monitoring for the detection of occult paroxysmal atrial fibrillation in ischemic stroke of unknown cause. Circulation 2011; 124(4): Huber K, Connolly SJ, Kher A et al. Practical use of dabigatran etexilate for stroke prevention in atrial fibrillation. Int J Clin Pract 2013; 67(6): Harris K, Mant J. Potential impact of new oral anticoagulants on the management of atrial fibrillation-related stroke in primary care. Int J Clin Pract 2013; 67(7): Sinha AM, Diener HC, Morillo CA et al. Cryptogenic Stroke and underlying Atrial Fibrillation (CRYSTAL AF): design and rationale. Am Heart J 2010; 160(1): e1. 59 Cotter PE, Martin PJ, Ring L, Warburton EA, Belham M, Pugh PJ. Incidence of atrial fibrillation detected by implantable loop recorders in unexplained stroke. Neurology 2013; 80(17): Ritter MA, Kochh auser S, Duning T et al. Occult atrial fibrillation in cryptogenic stroke: detection by 7-day electrocardiogram versus implantable cardiac monitors. Stroke 2013; 44(5): Etgen T, Hochreiter M, Mundel M, Freudenberger T. Insertable cardiac event recorder in detection of atrial fibrillation after cryptogenic stroke: an audit report Stroke 2013; 44(7): Paper received July 2013, accepted September 2013