Elsevier

Heart Rhythm

Volume 3, Issue 3, March 2006, Pages 296-310
Heart Rhythm

Original-clinical
Electrocardiographic imaging of cardiac resynchronization therapy in heart failure: Observation of variable electrophysiologic responses

https://doi.org/10.1016/j.hrthm.2005.11.025Get rights and content

Background

Cardiac resynchronization therapy (CRT) for congestive heart failure patients with delayed left ventricular (LV) conduction is clinically beneficial in approximately 70% of patients. Unresolved issues include patient selection, lead placement, and efficacy of LV pacing alone. Being an electrical approach, detailed electrical information during CRT is critical to resolving these issues. However, electrical data from patients have been limited because of the requirement for invasive mapping.

Objectives

The purpose of this study was to provide observations and insights on the variable electrophysiologic responses of the heart to CRT using electrocardiographic imaging (ECGI).

Methods

ECGI is a novel modality for noninvasive epicardial mapping. ECGI was conducted in eight patients undergoing CRT during native rhythm and various pacing modes.

Results

In native rhythm (six patients), ventricular activation was heterogeneous, with latest activation in the lateral LV base in three patients and in the anterolateral, midlateral, or inferior LV in the remainder of patients. Anterior LV was susceptible to block and slow conduction. Right ventricular pacing improved electrical synchrony in two of six patients. LV pacing in three of four patients involved fusion with intrinsic excitation resulting in electrical resynchronization similar to biventricular pacing. Although generally electrical synchrony improved significantly with biventricular pacing, it was not always accompanied by clinical benefit.

Conclusion

Results suggest that (1) when accompanied by fusion, LV pacing alone can be as effective as biventricular pacing for electrical resynchronization; (2) right ventricular pacing is not effective for resynchronization; and (3) efficacy of CRT depends strongly on the patient-specific electrophysiologic substrate.

Introduction

Delayed ventricular activation,1 predominantly of the left ventricle (LV), accompanies more than 30% of advanced heart failure cases. Cardiac resynchronization therapy (CRT) using biventricular pacing has been developed to restore synchrony. CRT improves patients’ symptoms and echocardiographic measures, and it confers a mortality benefit.2 However, many issues related to CRT remain unresolved. For reasons that are not clear, approximately 33% of patients do not respond to CRT [Multicenter InSync Randomized Clinical Evaluation (MIRACLE) trial].3 QRS duration (QRSd) as a criterion for patient selection and assessment of response is weak.4, 5 LV lead location may be a factor. The incremental benefit of single-site LV pacing alone is unresolved. The electrical consequences of LV pathology have not been examined.

Although improved mechanical performance is the aim of therapy, CRT is an electrical approach that uses pacing and electrophysiologic principles to modify mechanical behavior. The efficacy of the therapy is initially governed by electrical properties and thus may be limited by electrical dysfunction. Hence, failure to achieve resynchronization could be due, in part, to abnormal local electrical properties, such as slow conduction or conduction block, which are likely to be present in patients with LV disease. Therefore, an understanding of the electrophysiologic effects of CRT is essential, requiring high-resolution electrical mapping of ventricular excitation. Until recently, such information could not be obtained noninvasively.

Electrocardiographic imaging (ECGI) is a novel noninvasive imaging modality developed in our laboratory for determining cardiac excitation through noninvasive electrophysiologic epicardial imaging during a single beat. The method has been validated extensively in a human-shaped torso tank containing normal and abnormal canine hearts6, 7, 8 and in canine experiments,9, 10 demonstrating high accuracy in imaging focal activity, infarct substrate, repolarization gradients, and reentry during ventricular tachycardia. ECGI has been applied in humans11 and compared to direct epicardial mapping during open heart surgery12 and to catheter mapping during ventricular tachycardia.13 In humans, ECGI provided images of cardiac activation and repolarization during normal sinus rhythm, conduction disturbances, atrial flutter, and multisite ventricular pacing. A reconstruction accuracy better than 10 mm was consistently obtained in human subjects.14 Similar to previous studies, we used the ECGI-reconstructed epicardial data in this study to estimate the locations of pacing leads and compared them to the corresponding locations determined from computed tomographic (CT) images. Average accuracy within 10.2 ± 5.1 mm was achieved in locating the pacing sites. With this level of accuracy and extensive validation, ECGI provides a unique tool for noninvasive evaluation of electrophysiologic phenomena, including mechanisms and responses to therapeutic interventions.

In this study, ECGI was applied to eight patients with an implanted CRT device, and ventricular activation during native rhythm and various pacing modes was imaged. Our ultimate goal is to understand the electrophysiologic mechanism(s) that affects responses to CRT and to identify possible use of ECGI in guiding patient selection and lead placement.

This initial study provides preliminary but unique observations on the electrical activity of the heart during CRT and identifies electrical properties that may influence its success as a therapeutic approach. We posed the following hypothesis-driven questions: (1) Is native activation pattern in heart failure patients with left bundle branch block (LBBB) similar in all subjects? (2) Does LV pacing alone evoke similar electrical synchrony as biventricular pacing? (3) Does electrical synchrony correlate with clinical response? (4) Is right ventricular (RV) pacing alone beneficial? These questions were tested in a group of patients with severe LV disease exposed to various pacing modes. The results demonstrated several variables that likely significantly affect pacing efficacy.

Section snippets

Methods

The study evaluated eight heart failure patients (six men and two women; age 72 ± 11 years; New York Heart Association functional class III–IV) with ventricular conduction delay who received CRT through implanted atrial-biventricular pacing device. Six patients had native rhythm with LBBB pattern, one had atrial fibrillation, and one had no underlying rhythm (no AV conduction and no ventricular escape beats) at the time of the study (Table 1). The last two patients were upgraded to

Native rhythm

Native rhythm was imaged in six patients by inactivating the CRT device. Four representative maps are shown in Figure 1. ECG was similar to the preimplant ECG, indicating similar activation sequences. In all six patients, RV epicardial activation started from RV breakthrough 29 ± 7 ms after onset of body surface QRS. This is delayed relative to that in normal young adults.16 Epicardial activation spread radially from the RV breakthrough site, with latest activation of the RV at basal RV or

Discussion

The unique capability of ECGI for noninvasive epicardial mapping was used to determine activation sequences during pacing in a group of eight patients undergoing CRT with the following observations. (1) Activation sequences during native rhythm were heterogeneous among individuals. (2) RV pacing was not effective in resynchronization in most patients. (3) LV pacing often generated similar electrical synchrony as biventricular pacing due to fusion. (4) Anterior LV pacing was less effective than

Acknowledgments

We thank Paul Kohanski, Leslie Ciancibello, Sheila Shaffer, and Elena DuPont for assistance in pacing and CT data acquisition and transfer.

References (26)

  • H.S. Oster et al.

    Noninvasive electrocardiographic imagingreconstruction of epicardial potentials, electrograms, and isochrones and localization of single and multiple electrocardiac events

    Circulation

    (1997)
  • J.E. Burnes et al.

    Noninvasive ECG imaging of electrophysiologically abnormal substrates in infarcted heartsa model study

    Circulation

    (2000)
  • J.E. Burnes et al.

    Imaging dispersion of myocardial repolarization, Icomparison of body-surface and epicardial measures

    Circulation

    (2001)
  • Cited by (157)

    View all citing articles on Scopus

    The study was supported by Merit Award R37-HL-33343 and Grant R01-HL-49054 from the National Heart, Lung, and Blood Institute of the National Institutes of Health to Dr. Rudy. Dr. Rudy is the Fred Saigh Distinguished Professor at Washington University in St. Louis. Drs. Jia and Ramanathan are employees and stockholders of Cardioinsight, Inc., a company that plans to commercialize electrocardiographic imaging as a clinical tool. Dr. Rudy is a coinventor of the ECGI technology.

    View full text