This proposal describes an ancillary study to a recently initiated comparative effectiveness trial """"""""Surgical Ablation versus No Surgical Ablation for Patients with Persistent or Longstanding Persistent Atrial Fibrillation Undergoing Mitral Valve Surgery"""""""" (NCT00903370), which is part of the NHLBI-sponsored Cardiothoracic Surgery Network (CTSN).
Its aims are to perform a time-sensitive substudy of continuous heart rhythm monitoring and to develop and apply novel statistical tools to assess ablation effectiveness in groups of patients. As such, it will provide critical new knowledge to support a secondary objective of the parent trial: to inform rhythm monitoring and analysis strategies for future trials of rhythm control in atrial fibrillation (AF). Great enthusiasm for innovative tissue-destroying (ablation) procedures to treat AF has led to technology outstripping science. Mechanisms of AF (which afflicts >3 million Americans) remain poorly understood, methods to assess its cure are primitive, and guidelines for analyzing rhythm outcomes are flawed due to lack of appropriate statistical methodology. Why? After ablation procedures, AF may recur in episodes of varying duration. Yet current monitoring consists of """"""""snapshots"""""""" of semiannual rhythm recording (Holter), or more frequent-but short-transtelephonic monitoring (TTM). Ideally, an implantable monitoring device could detect all these episodes over a patient's lifetime;however, even if this were possible, statistical methods to analyze this series of occurrences in groups of patients for comparative effectiveness of AF therapies or to identify risk factors are poorly developed. Right now a window of opportunity exists to make important strides in more thorough AF assessment and proper data analysis because the CTSN has just launched a randomized trial of surgical AF therapies that includes weekly TTM of atrial rhythm as well as traditional 6- and 12-month Holter assessment. We propose a time-sensitive substudy of implanting an FDA-approved miniature loop recorder to capture AF episodes continuously. These data will be used in the development of a suite of new analytic methods for aggregating data such as these across time and across patients to characterize time course and identify modulating factors. They will then be used to assess the trial's rhythm endpoints and suggest designs for more informative future trials with substantially reduced sample size. We have assembled a team of clinicians and statisticians, including individuals who designed the parent trial, who are experienced in implanting the monitoring device, and who are experts in analytic methods research, to develop and apply novel methods to these several types of rhythm data. We will leverage developments from an intramural NHLBI machine learning program at Cleveland Clinic and State of Ohio-funded Atrial Fibrillation Innovation Center. The analytic methods have wide applicability to the important field of longitudinal data analysis.

Public Health Relevance

A chaotic heart rhythm called atrial fibrillation (AF), which affects more than 3 million Americans, especially the elderly, causes strokes, and commonly accompanies heart valve disease. AF treatments outstrip scientific understanding of the rhythm and how to assess treatment success across time in clinical trials. Thus, the Institute of Medicine has placed trials of AF treatment in the first tier of 100 comparative effectiveness priorities for the nation. An NIH-sponsored trial of surgical treatment of AF is has just been launched, and we propose a substudy of an implanted rhythm monitoring device (Medtronic REVEAL XT) and new methods of AF analysis for groups of patients that will help assess the trial's endpoints and inform future trials. Thus, this ancillary study application proposes to (1) collect, in a subset of patients, more refined rhythm data than proposed in the parent trial, (2) develop generalizable new methods of analysis, and (3) apply these methods to both the data collected through continuous monitoring as well as the follow-up data collected in the parent trial through Holter and trans-telephonic monitoring. (End of Abstract)

National Institute of Health (NIH)
National Heart, Lung, and Blood Institute (NHLBI)
Research Project (R01)
Project #
Application #
Study Section
Special Emphasis Panel (ZHL1)
Program Officer
Miller, Marissa A
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Cleveland Clinic Lerner
Schools of Medicine
United States
Zip Code
Pande, Amol; Li, Liang; Rajeswaran, Jeevanantham et al. (2017) Boosted Multivariate Trees for Longitudinal Data. Mach Learn 106:277-305
Rajeswaran, Jeevanantham; Blackstone, Eugene H (2017) A multiphase non-linear mixed effects model: An application to spirometry after lung transplantation. Stat Methods Med Res 26:21-42
Li, Liang; Mao, Huzhang; Ishwaran, Hemant et al. (2017) Estimating the prevalence of atrial fibrillation from a three-class mixture model for repeated diagnoses. Biom J 59:331-343
Suri, Rakesh M; Minha, Sa'ar; Alli, Oluseun et al. (2016) Learning curves for transapical transcatheter aortic valve replacement in the PARTNER-I trial: Technical performance, success, and safety. J Thorac Cardiovasc Surg 152:773-780.e14
Rajeswaran, Jeevanantham; Blackstone, Eugene H; Barnard, John (2016) Evolution of association between renal and liver functions while awaiting heart transplant: An application using a bivariate multiphase nonlinear mixed effects model. Stat Methods Med Res :
Minha, Sa'ar; Waksman, Ron; Satler, Lowell P et al. (2016) Learning curves for transfemoral transcatheter aortic valve replacement in the PARTNER-I trial: Success and safety. Catheter Cardiovasc Interv 87:165-75
Svensson, Lars G; Pillai, Saila T; Rajeswaran, Jeevanantham et al. (2016) Long-term survival, valve durability, and reoperation for 4 aortic root procedures combined with ascending aorta replacement. J Thorac Cardiovasc Surg 151:764-774.e4
Alli, Oluseun; Rihal, Charanjit S; Suri, Rakesh M et al. (2016) Learning curves for transfemoral transcatheter aortic valve replacement in the PARTNER-I trial: Technical performance. Catheter Cardiovasc Interv 87:154-62
Smedira, Nicholas G; Blackstone, Eugene H; Ehrlinger, John et al. (2015) Current risks of HeartMate II pump thrombosis: Non-parametric analysis of Interagency Registry for Mechanically Assisted Circulatory Support data. J Heart Lung Transplant 34:1527-34
Johnston, Douglas R; Soltesz, Edward G; Vakil, Nakul et al. (2015) Long-term durability of bioprosthetic aortic valves: implications from 12,569 implants. Ann Thorac Surg 99:1239-47

Showing the most recent 10 out of 12 publications