Using Remote Monitoring to Guide Anticoagulation Decisions

Among the most common clinical issues we face as electrophysiologists involve decisions surrounding the initiation and continuation of anticoagulation therapy for our patients with atrial fibrillation (AF). Unfortunately, some of our approaches are based on dated information that fails to leverage advances in technology and pharmacology; changes that potentially can alter these critical aspects of patient care. Remote monitoring of cardiac implantable electronic devices (CIEDs) may allow for the early recognition of AF and provide a window of opportunity for stroke prevention before the occurrence of a catastrophic cardioembolic event. This same technology may be used to target anticoagulation around the time of an AF episode in select individuals with a known AF history, thus limiting the exposure and inherent risks of lifelong anticoagulation without compromising stroke risk. Whether available evidence supports the use of remote monitoring to guide these decisions is the subject of ongoing debate.

Stroke remains the most devastating consequence of AF and is the first manifestation of the arrhythmia in a quarter of all AF-related strokes.¹ AF screening of the population would appear attractive at first glance, but is fraught with challenges. In general, we reserve screening for diseases that are serious, when treatment before event onset is more effective than treatment rendered after an adverse event occurs, and when the prevalence of the disease during the detectable pre-clinical phase is high. Given the growing incidence and prevalence of AF, and the marked reduction in cardioembolic events seen with anticoagulation, screening an “at risk” population for AF appears to make sense. However, the often paroxysmal and asymptomatic nature of AF, combined with the mounting evidence linking even brief episodes with stroke, makes screening problematic. The use of opportunistic screening (i.e., pulse check and ECG if abnormal) and short-term ECG monitoring have proven superior to routine care in detecting AF, but is clearly an underestimate of the true AF burden in the population.^2,3 Furthermore, whether such strategies reduce stroke risk has yet to be determined. In contrast, the continuous recording and remote transmission capabilities of CIEDs provide the best opportunity for early AF recognition and management in patients without a prior AF history. Observational studies of ICDs and pacemakers found that remote monitoring would detect AF 64 to 164 days earlier than in-office monitoring performed every 3 to 6 months, respectively.^4,5 Trials including TRUST, CONNECT, and COMPAS all demonstrated that remote monitoring of CIEDs reduced the time to AF detection by weeks to months depending on the expected frequency of routine follow-up.^6-8 In high-risk individuals such as those with cryptogenic stroke, the CRYSTAL-AF trial showed that continuous remote monitoring with an implantable cardiac monitor detected AF at a rate of 30% at 3 years in the intervention arm compared to only 3% in patients receiving routine follow-up care.^9,10

While it would seem obvious that early detection using remote monitoring would reduce the risk of stroke or recurrent stroke, this has yet to be proven. None of the above-referenced studies thoroughly evaluated the impact of early AF diagnosis on treatment decisions or outcomes. Although CRYSTAL-AF did show that prescription rates of anticoagulation once AF was discovered approached 90% at 3 years, the study was not powered to show a reduction in recurrent stroke risk, and the observed reduction in recurrent events in those randomized to the implantable cardiac monitor arm did not reach statistical significance. While early AF detection and treatment seems logical, the IMPACT trial is evidence that early detection of AF does not necessarily reduce outcomes in patients without a known history of AF.¹¹ IMPACT used frequent home monitoring of ICDs and CRT-Ds to initiate and terminate anticoagulation based on underlying stroke risk and AF duration. Though the trial showed no difference in the combined endpoint of stroke and major bleeds, a closer look raises some questions over the validity of its conclusions. In IMPACT, patients were started on oral anticoagulation, predominantly warfarin, if they had a threshold AF event of a duration that varied by CHADS₂ score. The decision to stop anticoagulation after a predefined period was also based on stroke risk, with higher risk patients requiring >90 days of freedom from atrial tachyarrhythmias prior to discontinuing anticoagulation. Overall compliance with the protocol was poor, and the study design allowed for several days to pass before anticoagulation was initiated. The use of warfarin as the main anticoagulant meant that several more days passed from the onset of an AF episode to the achievement of a therapeutic INR. The incidence of AF was much lower than expected, and the small reduction in stroke risk was offset by higher bleed rates seen in those patients treated with oral anticoagulation. Given these limitations, the question remains whether a positive result could have been seen if rapid onset novel oral anticoagulants were used, if the study protocol were adhered to, if the study was powered for a primary stroke endpoint, or if a population without an ICD indication (i.e., systolic heart failure) were targeted. The ongoing debate about the relationship between AF duration and stroke and its responsiveness to oral anticoagulation will be addressed by two trials: ARTESiA (Apixaban for the Reduction of Thrombo-Embolism in Patients With Device-Detected Sub-Clinical Atrial Fibrillation; clinicaltrials.gov identifier NCT01938248), and NOAH (Non-vitamin K Antagonist Oral Anticoagulants in Patients With Atrial High Rate Episodes; clinicaltrials.gov identifier NCT02618577). Until these studies are completed, the treatment of device-detected “subclinical” episodes of AF, particularly of short duration, remains controversial and may be best managed using careful consideration of AF duration and underlying stroke risk on a per-patient basis.

Perhaps the greatest utility of remote monitoring may not be in the early detection of AF in those with no history, but in the management of anticoagulation in those that already carry an AF diagnosis. These patients may be post ablation, managed with antiarrhythmic drugs alone, or have undergone cardioversion for one or more episodes of persistent AF. Current practice recommends that chronic anticoagulation be considered based on underlying stroke risk, despite the fact that sinus rhythm may be maintained forever in some patients, and AF may recur very infrequently in others (Figure 1). Clinical trial data comparing rhythm versus rate control have informed a generation of physicians on the need for chronic anticoagulation in these individuals, even in the face of presumed long-term sinus rhythm. The absence of any permanent cure for AF, high long-term recurrence rates following cardioversion alone, high burden of asymptomatic episodes, and the uncertainties surrounding the relationship between AF duration, timing, and stroke are among the strong arguments for why we have practiced in this manner. As a result, we currently have 2 options for treating patients with a “successful” rhythm control strategy: 1) continue anticoagulation indefinitely, thereby exposing the patient to the risks of a drug from which they are unlikely to derive any benefit during prolonged periods of sinus rhythm; or 2) stop OAC, thereby reducing the risk of bleeding but exposing the patient to the risk of stroke should AF recur. Remote monitoring of implantable cardiac devices has been used to discontinue anticoagulation in select patients with a rhythm control strategy. In one study of patients with elevated CHA₂DS₂-VASC and HAS-BLED scores, anticoagulation was stopped in those patients with an AF burden <1% on an implantable cardiac monitor without any neurologic sequela over a median follow-up time of 2 years.¹² This same technology was used to permanently discontinue anticoagulation in a post-ablation CHADS₂ 1-3 population without documented AF 3 months post procedure.¹³ During a follow-up time of 32 months, 63% had an AF burden of <1 hour/day and were able to stay off OAC. No strokes, TIAs, or other thromboembolic events were observed during follow-up. The ability to use daily remote monitoring of CIEDs combined with the use of a rapid acting novel oral anticoagulant that provides systemic anticoagulation within 1-4 hours of a single oral dose allows for a third option: withhold anticoagulation and target treatment only in response to an AF episode. This “pill in the pocket” approach to anticoagulation may limit the risks of lifelong exposure to anticoagulation without compromising stroke risk. Targeted therapy would almost certainly improve medication adherence during the period of highest risk, and prove to be cost-effective given the financial burden of chronic anticoagulation and the associated complications. In the Rhythm Evaluation for Anticoagulation Therapy with Continuous Monitoring (REACT.COM) pilot study (clinicaltrials.gov identifier NCT01706146), 59 patients with a CHADS₂ score of 1 or 2 were treated with a minimum of 30 days of a novel oral anticoagulant only in response to an episode of AF lasting >1 hour detected on an implantable cardiac monitor.¹⁴ Over a median follow-up time of 14 months, only 31% of patients had any threshold AF event and the observed reduction in “time on” anticoagulation was 94% without any observed strokes (Figure 2). There were 2 “possible” and 1 “definite” TIAs, all of which occurred in patients with a CHADS₂ score of 1 receiving aspirin 81 mg and all in the absence of a threshold AF event in the preceding >12 months (Figure 3). The TACTIC-AF trial (Pilot Study: Tailored Anticoagulation for Noncontinuous AF; clinicaltrials.gov identifier NCT01650298) uses a similar strategy in patients with dual-chamber pacemakers or ICDs, with a CHADS₂ score of 1-3 and no prior stroke history. AF episodes >30 minutes or a total daily burden of AF >6 hours results in a month of anticoagulation with a NOAC.¹⁵ This single-arm pilot study has completed enrollment and is expected to complete follow-up in 2016. Clearly, whether such an approach can ever be adopted for widespread use will depend on both a pivotal trial comparing chronic anticoagulation versus CIED-guided “pill in the pocket” anticoagulation and either the development of an infrastructure capable of intensively monitoring a large number of patients or advances in wireless technology that permit real-time communication between CIEDs and smartphones to allow for point-of-care, patient-directed therapy. 

If “pill in the pocket” anticoagulation is ever embraced by the electrophysiology community, it will require us to address a basic knowledge gap in the temporal association between AF and stroke. Though the Framingham Heart Study showed us a fivefold increase in stroke risk in patients with non-valvular AF, the temporal association between AF and stroke is controversial. Studies including ASSERT, TRENDS, and IMPACT show that many strokes occur months after the last episode of AF or in the absence of AF altogether, leading some to believe that AF is simply a marker for stroke and not in the causal pathway.^11,16,17 On the other hand, the largest study to date showed a fivefold increase in stroke risk during the 30 days after an AF event, supporting the concept that AF is the cause of stroke in many patients and not simply a marker for risk.¹⁸ The truth may lie somewhere in-between and may depend on a patient’s underlying stroke risk and the duration of AF. For example, a patient with a CHA₂DS₂-VASc score of 5 or more may have multiple stroke mechanisms and the risk of stroke due to AF may be only one of several potential sources. This is supported by the ASSERT trial, in which the 51 strokes that occurred in 2580 patients had a mean age of 78 and a mean CHA₂DS₂-VASc score of 4.5. Furthermore, the risk of stroke did not increase to a statistically significant degree until AF duration approached 18 hours.¹⁷ The temporal dissociation between AF and stroke can also be explained by the possibility that thrombi initiate during an AF episode, but propagate and embolize long after sinus rhythm has been restored. Whether prompt recognition and anticoagulation could prevent downstream thromboembolic events remains to be determined. However, a trial of targeted anticoagulation using CIEDs performed on patients on the lower end of the stroke-risk spectrum would definitely address this issue.

The advent of continuous remote monitoring has raised more questions about AF than it has answered. Ongoing studies will address the clinical risk of device-detected “subclinical” AF and its responsiveness to anticoagulation. In addition, planned studies will evaluate the potential for targeted anticoagulation guided by remote monitoring in a manner that was unimaginable just a few years ago. This personalized approach to anticoagulation has the potential to reduce bleeding, reduce costs, and improve quality of life to millions of patients with AF. Of course, we must first prove that these benefits do not occur at the cost of an increased stroke risk. If this targeted approach is proven safe and effective, it may also expand the indications for rhythm control beyond symptom relief to include the goal of reducing the exposure to long-term anticoagulation. As we move towards an era of personalized medicine, we should be questioning whether the way we practice today makes sense and continually ask ourselves how we can leverage the advances in remote monitoring within our own field to better care for our patients.

Disclosures: The authors have no conflicts of interest to report regarding the content herein. Outside the submitted work, Dr. Passman reports receiving a grant and personal fees from Medtronic and personal fees from Pfizer.  

In the next issue of EP Lab Digest, the topic of the Remote Healthcare section will be “Efficacy of Remote Monitoring as a Function of CIED Type” by Dr. Niraj Varma.

References

1. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke. 1991;22(8):983-988.
2. Fitzmaurice DA, Hobbs FD, Jowett S, et al. Screening versus routine practice in detection of atrial fibrillation in patients aged 65 or over: cluster randomised controlled trial. BMJ. 2007;335(7616):383.
3. Svennberg E, Engdahl J, Al-Khalili F, Friberg L, Frykman V, Rosenqvist M. Mass Screening for Untreated Atrial Fibrillation: The STROKESTOP Study. Circulation. 2015;131(25):2176-2184.
4. Lazarus A. Remote, wireless, ambulatory monitoring of implantable pacemakers, cardioverter defibrillators, and cardiac resynchronization therapy systems: analysis of a worldwide database. Pacing Clin Electrophysiol. 2007;30(Suppl 1):S2-S12.
5. Ricci RP, Morichelli L, Santini M. Remote control of implanted devices through Home Monitoring technology improves detection and clinical management of atrial fibrillation. Europace. 2009;11(1):54-61.
6. Varma N, Epstein AE, Irimpen A, Schweikert R, Love C; TRUST Investigators. Efficacy and safety of automatic remote monitoring for implantable cardioverter-defibrillator follow-up: the Lumos-T Safely Reduces Routine Office Device Follow-up (TRUST) trial. Circulation. 2010;122(4):325-332.
7. 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):1181-1189. 
8. Mabo P, Victor F, Bazin P, Ahres S, Babuty D, Da Costa A, Binet D, Daubert JC; COMPAS Trial Investigators. A randomized trial of long-term remote monitoring of pacemaker recipients (the COMPAS trial). Eur Heart J. 2012;33(9):1105-1111. 
9. Sanna T, Diener H, Passman R, et al, on behalf of CRYSTAL AF investigators. Cryptogenic Stroke and Underlying Atrial Fibrillation (CRYSTAL AF). N Engl J Med. 2014;370:2478-2486.
10. Brachmann J, Morillo CA, Sanna T, et al. Uncovering Atrial Fibrillation Beyond Short-Term Monitoring in Cryptogenic Stroke Patients: Three-Year Results From the Cryptogenic Stroke and Underlying Atrial Fibrillation Trial. Circ Arrhythm Electrophysiol. 2016;9(1):e003333.
11. Martin DT, Bersohn MM, Waldo AL, et al; IMPACT Investigators. Randomized trial of atrial arrhythmia monitoring to guide anticoagulation in patients with implanted defibrillator and cardiac resynchronization devices. Eur Heart J. 2015;36(26):1660-1668.
12. Mascarenhas DA, Farooq MU, Ziegler PD, Kantharia BK. Role of insertable cardiac monitors in anticoagulation therapy in patients with atrial fibrillation at high risk of bleeding. Europace. 2016;18(6):799-806. 
13. Zuern CS, Kilias A, Berlitz P, et al. Anticoagulation after catheter ablation of atrial fibrillation guided by implantable cardiac monitors. Pacing Clin Electrophysiol. 2015;38(6):688-693.
14. Passman R, Leong-Sit P, Andrei A, et al. Targeted Anticoagulation for Atrial Fibrillation Guided by Continuous Rhythm Assessment with an Insertable Cardiac Monitor: The Rhythm Evaluation for Anticoagulation with Continuous Monitoring (REACT.COM) Pilot Study. J Cardiovasc Electrophysiol. 2016;27(3):264-270.
15. Zimetbaum P, Ellis E, Waks J, Passman R. The Importance of Atrial Fibrillation Burden and the Origin of Device-Tailored Anticoagulation. Pacing Clin Electrophysiol. 2013;36(11):1319-1324.
16. 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):1416-1423. 
17. Brambatti M, Connolly SJ, Gold MR, et al; ASSERT Investigators. Temporal relationship between subclinical atrial fibrillation and embolic events. Circulation. 2014;129(21):2094-2099.
18. Turakhia MP, Ziegler PD, Schmitt SK, et al. Atrial Fibrillation Burden and Short-Term Risk of Stroke: Case-Crossover Analysis of Continuously Recorded Heart Rhythm From Cardiac Electronic Implanted Devices. Circ Arrhythm Electrophysiol. 2015;8(5):1040-1047.