Cardiac Rhythm Management
Articles Articles 2015 January

Expert Commentary From The Section Editor

DOI: 10.19102/icrm.2015.060104

Rahul N. Doshi, MD, FHRS, FACC

PDF Download PDF


Atrial Fibrillation Detected via Remote Monitoring: Big Data Solutions for Big Problems--or just more questions?

"You can have data without information, but you cannot have information without data." Daniel Keys Moran, computer programmer and science fiction author.

“Information is the oil of the 21st century, and analytics is the combustion engine.” Peter Sondergaard, Gartner Research.

Atrial fibrillation (AF) is the most prevalent arrhythmia in the world, and is the most costly arrhythmia whether measured by patient outcomes or financial burden to our healthcare system. Similarly, heart failure (HF) is the most costly diagnosis for all of healthcare. It would thus stand to reason that AF in the HF population is a significant problem. While estimates of AF in the general population is approximately 5%, in patients with heart failure the prevalence may be as high as 40%1,2. Understanding the impact of AF within the HF population has significant implications spanning the spectrum from our individual patient management decisions to global healthcare policy.

In this issue of the Journal, Cesario et al on behalf of the ALTITUDE Study Group examine the effects of the presence of AF on shock-burden and overall mortality in a large population of patients with CRT-D devices followed via remote monitoring. Pulling from a study population at least an order of magnitude greater than the traditional CIEDS trial, they report a high burden of AF in this population of 47.1%. The authors demonstrate that increasing AF burden and AF duration are associated with worsening outcomes, and that the increase in mortality is across a broad spectrum of both burden and duration. Thus, AF that previously may have been thought of as trivial is associated with a poorer prognosis in this population of CRT-D patients.

In addition, the authors suggest that inappropriate shocks for AF are not responsible for worsening outcomes but suggest that it is the presence of AF itself. The presence of AF was found to be a more powerful predictor for adverse outcomes than a inappropriate shock for AF. Moreover, conversion of AF episodes with ICD therapy did not result in a favorable outcome compared to subjects in which AF failed to convert the arrhythmia. These observations support the notion that it is the presence of AF that is associated with a worsened outcome in CRT-D recipients and that this may be a marker of disease severity or clinical deterioration.

While it is generally well established that AF is associated with worsening outcomes in the HF population, the power of this report and other observations from the ALTITUDE Study Group should not be discounted3. These particular observations are made from a study population of almost 64,000 patients with over a 3-year follow-up. The authors point out that the power of this data set is not just the large population but also the nature of how patients were characterized. Compared to prior studies in the ICD population in which the presence of AF was made at the time of enrollment, remote monitoring evaluates patients continuously and thus might be more reflective of an ever-changing disease and substrate in this population.

There are now increasing examples how CIEDS-based rhythm diagnostics may lead to better understanding of disease entities and lead to treatment strategies previously overlooked. The power of the ASSERT study is not just based on the accuracy of device detection algorithms but in the demonstration that asymptomatic episodes of AF increase the risk of thromboembolism4. Episodes of AF that would have been previously undetected are associated with a tangible risk that may influence how we treat our patients. Cesario and colleagues make an important observation regarding the risk that AF conveys in this population of CRT-D recipients, but much like the ASSERT study do not answer the question of whether or not therapy based on these diagnostics will lead to better patient outcomes or what kind of therapy should we initiate.

CIEDS-based diagnostics are only one example of technology solutions that are more patient-centered and continuous in nature and represent a part of the growing field in medicine known as Body Computing. Implantable or wearable devices that are networked and evaluated in real time have growing implications in how we will be evaluating and treating patients. Clinical cardiac electrophysiology is a technology-based subspecialty and is uniquely situated to utilize such networks5. The information generated from such technology, much like the manuscript from Cesario et al, have the potential to allow for novel insights regarding these disease entities.

But do these “big-data” modalities generate solutions or simply more questions? We know that AF is associated with worsening outcomes in the HF population, but is it causative? Does AF increase HF in this population by changes in mechanical function, neurohormonal changes, loss of CRT therapy or other mechanisms? Or is it simply a by-product of a progressive disease? More importantly, if we treat AF, can we modify the disease progression and lead to improved outcomes?

Thus far, the benefits of treating AF in the HF population are far from definitive. The AF-CHF trial sought to evaluate the effects of a rhythm control strategy or maintenance of sinus rhythm in a HF population with low ejection fraction, and did not demonstrate an improved outcome in either setting6. However, the type of treatment strategy may be important in this population. Much like the wealth of data demonstrated in the post-AFFIRM era, any potential benefit of rhythm control may be negated by pharmacologic approaches. Curative ablation may represent a more effective means of treating AF and may modify or reverse disease progression7. As the data is limited, regardless of the treatment strategy, we still do not know whether or not we are treating an epi-phenomenon or a critical component of disease progression.

Remote monitoring combined with novel sensor technology may provide further insights to these questions. Impedance-based measurements using existing device technology can assess lung water volume and hence pulmonary edema. More recently, pressure-sensor technology such as pulmonary arterial or left atrial pressure recordings can be followed as traditional CIEDS. These devices may be able to definitively answer the question of whether hemodynamic alterations lead to AF or vice versa. This may significantly alter or determine our course of action in treating this particular population. Will we need to more aggressively treat the hemodynamic or volume status of a patient with HF, which will result in less AF? Or, conversely, do we need to aggressively treat AF in the HF population to prevent the hemodynamic consequences that AF precipitates?

In this issue of the Journal, Cesario et al have demonstrated utilizing “big-data” that AF is associated with poor outcomes in a large population of CRT-D recipients. Big problems generate big data, and require big solutions. This does not just pertain to IBM anymore! We all can look forward to “big-data solutions” that answer fundamental questions critical to furthering both our understanding and the development of treatment strategies that will benefit our patient populations.

Rahul N. Doshi, MD, FHRS, FACC
Director, Clinical Cardiac Electrophysiology
Associate Professor of Medicine
Keck USC School of Medicine
Los Angeles, CA, USA


  1. DePasquale EC, Fonarow GC. Impact of Atrial Fibrillation on Outcomes in Heart Failure. Heart Failure Clin 9 (2013):437–449. [CrossRef] [PubMed]
  2. Wang TJ, Larson MG, Levy D, et al. Temporal relations of atrial fibrillation and congestive heart failure and their joint influence on mortality. Circulation 2003;107:2920–5. [CrossRef] [PubMed]
  3. Saxon LA, Hayes DL, Gilliam FR, et al. Long-term outcome After ICD and CRT implantation and influence of remote device follow-up: The ALTITUDE survival study. Circulation 2010; 122:2359–2367. [CrossRef] [PubMed]
  4. Healy JS, Connolly SJ, Gold MR, et al. Subclinical Atrial Fibrillation and the Risk of Stroke. N Engl J Med 2012;366:120-9. [CrossRef] [PubMed]
  5. Saxon LA, Rao AK. Body computing: How networked medical devices can solve problems facing health care today. J Cardiovasc Electrophysiol 2007; 18:1345–1347. [CrossRef] [PubMed]
  6. Talajic M, Khairy P, Levesque S, et al. Maintenance of Sinus Rhythm and Survival in Patients with Heart Failure and Atrial Fibrillation. J Am Coll Cardiol 2010; 55:1796–802. [CrossRef] [PubMed]
  7. Chen MS, Marrouche NF, Khaykin Y, et al. Pulmonary Vein Isolation for the Treatment of Atrial Fibrillation in Patients with Impaired Systolic Function. J Am Coll Cardiol 2004; 43:1004–9. [CrossRef] [PubMed]