Noninvasive electrocardiographic imaging (ECGI): comparison to intraoperative mapping in patients

Abstract

Objectives/background: Cardiac arrhythmias are a leading cause of death and disability. Electrocardiographic imaging (ECGI) is a noninvasive imaging modality that reconstructs potentials, electrograms, and isochrones on the epicardial surface from body surface measurements. We previously demonstrated in animal experiments through comparison with simultaneously measured epicardial data the high accuracy of ECGI in imaging cardiac electrical events. Here, images obtained by noninvasive ECGI are compared to invasive direct epicardial mapping in open heart surgery patients.

Methods: Three patients were studied during sinus rhythm and right ventricular endocardial and epicardial pacing (total of five datasets). Body surface potentials were acquired preoperatively or postoperatively using a 224-electrode vest. Heart-torso geometry was determined preoperatively using computed tomography. Intraoperative mapping was performed with two 100-electrode epicardial patches.

Results: Noninvasive potential maps captured epicardial breakthrough sites and reflected general activation and repolarization patterns, localized pacing sites to approximately 1 cm and distinguished between epicardial and endocardial origin of activation. Noninvasively reconstructed electrogram morphologies correlated moderately with their invasive counterparts (cross correlation = 0.72 +/- 0.25 [sinus rhythm], 0.67 +/- 0.23 [endocardial pacing], 0.71 +/- 0.21 [epicardial pacing]). Noninvasive isochrones captured the sites of earliest activation, areas of slow conduction, and the general excitation pattern.

Conclusions: Despite limitations due to nonsimultaneous acquisition of the surgical and noninvasive data under different conditions, the study demonstrates that ECGI can capture important features of cardiac electrical excitation in humans noninvasively during a single beat. It also shows that general excitation patterns and electrogram morphologies are largely preserved in open chest conditions.

Figure 1

Study design. A: Cross-sectional CT slice showing the heart and body surface electrodes. B: Vest for body surface potential mapping. C: Subject-specific heart-torso geometry digitized from CT (blue circles indicate electrode positions). D: Noninvasive epicardial potentials obtained from CT and body surface potential mapping data using electrocardiographic imaging. E: Epicardial patches used for intraoperative mapping. F: Snapshot of the intraoperative mapping procedure showing the anterior right ventricle (RV) with inserted epicardial patch and epicardial pacing wires.

Figure 2

Invasive (left column) and noninvasively imaged (middle column) epicardial potential maps from body surface potential mapping (BSPM; right column) 10 ms (top row), 21 ms (second row), and 30 ms (bottom row) following QRS onset. Numbers (second row) indicate right ventricular breakthrough sites. Left anterior descending coronary artery (LAD) and AV groove are marked in thick gray lines. LV = left ventricle.

Figure 3

Invasive (red) and noninvasively reconstructed (black) QRS electrograms with cross-correlation (CC) and time lag (in milliseconds) during sinus rhythm. A: Anterior right ventricle (RV). B: Posterior left ventricle (LV). Boundaries of epicardial patches are shown. Numbers identify electrogram locations. LAD = Left anterior descending coronary artery.

Figure 4

A: Invasive (first and last columns) and noninvasively imaged (second and third columns) epicardial potentials during right ventricular (RV) endocardial pacing. Times are relative to onset of pacing. Asterisk denotes pacing site. B: Invasive and noninvasively imaged potentials during repolarization (T peak). C: Invasive (red) and noninvasively reconstructed (black) electrograms. LV = left ventricle.

Figure 5

A: Patient OR3. Invasive potentials from lateral (first column) and anterior (last column) right ventricle (RV) with corresponding noninvasively imaged potentials (second and third columns), 80 ms (top row) following pacing and during repolarization (T peak, bottom row). Minus sign (−) denotes pacing site. B: Patient OR2. Invasive and noninvasively imaged epicardial potentials at 70 ms following pacing and during repolarization. Same format as panel A. LAD = Left anterior descending coronary artery.

Figure 6

Invasive (red) and noninvasively reconstructed (black) electrograms with cross-correlation (CC) and time lag (in milliseconds) during right ventricular (RV) epicardial pacing. A: Anterior RV. Asterisk denotes reconstructed pacing site. B: Lateral RV.

Figure 7

Invasive (first and last columns) and noninvasively imaged (second and third columns) isochrones. A: Sinus rhythm. Numbers indicate right ventricular (RV) breakthrough sites. B: RV endocardial pacing. C: RV epicardial pacing. Arrows point to regions of slow conduction (crowded isochrones). Asterisk denotes pacing site. LAD = left anterior descending coronary artery; LV = left ventricle.