8 - Randomized Clinical Trial of CMR-Guided Cardiac Resynchronization Therapy: Implant Findings

Saturday, February 15, 2020

2:25 PM – 2:35 PM

Location: Canary 3 & 4

Background: Response rates to cardiac resynchronization therapy (CRT) in patients with heart failure continue to remain suboptimal. This has highlighted the unmet need for strategic interventions using CMR to improve response rates. To this end, we have developed methods to determine mechanical activation from CMR Displacement Encoding with Stimulated Echoes (DENSE) and advanced co-registration methods for activation, scar, and coronary venous anatomy.

Methods: The inclusion and exclusion criteria are described on clinicaltrials.gov (NCT03398369). Briefly, all patients must have an indication for a de novo or upgraded CRT defibrillator. In intervention (CMR-guided) patients, a co-registered 3D map of CMR DENSE mechanical activation and scar is integrated with coronary venous anatomy from computed tomography angiography (CTA) to determine the optimal left ventricular pacing location prior to the procedure (Figure 1; optimal pacing site circled). A left ventricular (LV) pacing site with late activation and no scar is then targeted. Control patients also have a CMR prior to CRT, but the information is only used as baseline data for comparison with a 6-month post-CRT CMR. The primary endpoint is improvement in left ventricular function 6 months after CRT. In addition, patients have echocardiography, complete a heart failure questionnaire, and undergo cardiopulmonary exercise testing to assess peak VO₂ at baseline, 6 months post-CRT, and 12 months post-CRT.

Results: The 32 patients enrolled were 66.2 ÷ 10.8 years old, 31.3% were female, 25% were African American, and 44% had ischemic cardiomyopathy. Those randomized to CMR guidance for the CRT procedure had very favorable implant results. As shown in Figure 2, the QRS duration decreased by a median of 24 ms (IQR 10-47 ms) in CMR-guided patients but remained the same in controls (p=0.02) (A) with even more pronounced effects in patients with nonischemic cardiomyopathy (NICM) (p=0.007) (B). Figure 3 (A) demonstrates more favorable (greater) QRS-LV-electrogram (Q-LV) times at LV pacing sites in CMR-guided patients with the Q-LV:QRS ratio (expressed as a percentage) approaching unity in many (p=0.001 versus controls; latest activation time not in scar substituted for QRS duration in 2 patients). The differences were again even greater in NICM (p=0.004) (B). All LV pacing sites in CMR-guided patients were placed as intended in areas without scar, while 37.5% of controls had LV lead placement in scar (p=0.02 for difference in all patients; p=0.004 for difference in ischemic cardiomyopathy [ICM] only). Overall, no adverse events have been reported.

Conclusion: In this randomized clinical trial of CMR-guided CRT, we have demonstrated superior findings in CMR-guided patients, including greater QRS shortening after CRT and more favorable pacing site findings such as greater Q-LV times and absence of scar at LV pacing sites. We are continuing to enroll patients and are performing follow-up visits for enrolled patients to evaluate clinical endpoints.

Presenting Author(s)

Kenneth Bilchick, MD, MSc

Associate Professor
University of Virginia Health System

Disclosure: Medtronic (Individual(s) Involved: Self, Products/Services: Research Grant): Grant/Research Support; Siemens Healthineers (Individual(s) Involved: Self, Products/Services: Research Grant): Grant/Research Support

Primary Author(s)

Kenneth Bilchick, MD, MSc

Associate Professor
University of Virginia Health System

Disclosure: Medtronic (Individual(s) Involved: Self, Products/Services: Research Grant): Grant/Research Support; Siemens Healthineers (Individual(s) Involved: Self, Products/Services: Research Grant): Grant/Research Support

Co-Author(s)

Mohammad Abdishektaei, MSc

Graduate Student
University of Virginia Health System

Disclosure: Disclosure information not submitted.
TP

Thien-Khoi Phung, BEng

University of Virginia Health System

Disclosure: Disclosure information not submitted.
Daniel Auger, PhD

Client Services Manager / Research Scientist
Medical Metrics, Inc. / University of Virginia

Disclosure: Disclosure information not submitted.
XG

Xu Gao, MD

University of Virginia Health System

Disclosure: Disclosure information not submitted.
RM

Rohit Malhotra, MD

Associate Professor of Medicine
University of Virginia Health System

Disclosure: Disclosure information not submitted.
PM

Pamela Mason, MD

Associate Professor of Medicine
University of Virginia Health System

Disclosure: Disclosure information not submitted.
NM

Nishaki Mehta, MD

University of Virginia Health System

Disclosure: Disclosure information not submitted.
AD

Andrew Darby, MD

Associate Professor of Medicine
University of Virginia Health System

Disclosure: Disclosure information not submitted.
MM

Michael Mangrum, MD

Associate Professor of Medicine
University of Virginia Health System

Disclosure: Disclosure information not submitted.
Christopher Kramer, MD

Ruth C. Heede Professor of Cardiovascular Medicine
University of Virginia

Disclosure: Disclosure information not submitted.
JH

Jeffrey Holmes, MD, PhD

University of Virginia Health System

Disclosure: Disclosure information not submitted.
Frederick Epstein, PhD

Mac Wade Professor
University of Virginia

Disclosure: Disclosure information not submitted.