Non-invasive three-dimensional electrical activation mapping to predict cardiac resynchronization therapy response: site of latest left ventricular activation relative to pacing site.

AIMS: Pacing remote from the latest electrically activated site (LEAS) in the left ventricle (LV) may diminish response to cardiac resynchronization therapy (CRT). We tested whether proximity of LV pacing site (LVPS) to LEAS, determined by non-invasive three-dimensional electrical activation mapping [electrocardiographic Imaging (ECGI)], increased likelihood of CRT response. METHODS AND RESULTS: Consecutive CRT patients underwent ECGI and chest/heart computed tomography 6-24 months of post-implant. Latest electrically activated site and the distance to LVPS (dp) were assessed. Left ventricular end-systolic volume (LVESV) reduction of ≥15% at clinical follow-up defined response. Logistic regression probabilistically modelled non-response; variables included demographics, heart failure classification, left bundle branch block (LBBB), ischaemic heart disease (IHD), atrial fibrillation, QRS duration, baseline ejection fraction (EF) and LVESV, comorbidities, use of CRT optimization algorithm, angiotensin-converting enzyme inhibitor(ACE)/angiotensin-receptor blocker (ARB), beta-blocker, diuretics, and dp. Of 111 studied patients [64 ± 11 years, EF 28 ± 6%, implant duration 12 ± 5 months (mean ± SD), 98% had LBBB, 38% IHD], 67% responded at 10 ± 3 months post CRT-implant. Latest electrically activated sites were outside the mid-to-basal lateral segments in 35% of the patients. dp was 42 ± 23 mm [31 ± 14 mm for responders vs. 63 ± 24 mm non-responders (P < 0.001)]. Longer dp and the lack of use of CRT optimization algorithm were the only independent predictors of non-response [area under the curve (AUC) 0.906]. dp of 47 mm delineated responders and non-responders (AUC 0.931). CONCLUSION: The distance between LV pacing site and latest electrical activation is a strong independent predictor for CRT response. Non-invasive electrical evaluation to characterize intrinsic activation and guide LV lead deployment may improve CRT efficacy.

Electrocardiographic imaging to guide ablation of ventricular arrhythmias and agreement between two different systems.

BACKGROUND AND AIM: A recent study using an epicardial-only electrocardiographic imaging (ECGI), suggests that the agreement of ECGI activation mapping and that of the contact mapping for ventricular arrhythmias (VA) is poor. The aim of this study was to assess the diagnostic value of two endo-epicardial ECGI systems using different cardiac sources and the agreement between them. METHODS: We performed 69 ECGI procedures in 52 patients referred for ablation of VA at our center. One system based on the extracellular potentials was used in 26 patients, the other based on the equivalent double layer model in 9, and both in 17 patients. The first uses up to 224 leads and the second just the 12­lead ECG. The localization of the VA was done using a segmental model of the ventricles. A perfect match (PM) was defined as a predicted location within the same anatomic segment, whereas a near match (NM) as a predicted location within the same segment or a contiguous one. RESULTS: 44 patients underwent ablation, corresponding to 58 ECGI procedures (37 with the first and 21 with the second system). The percentage of PMs and NMs was not significantly different between the two systems, respectively 76% and 95%, p = 0.077, and 97% and 100%, p = 1.000. In 14 patients that underwent ablation and had the ECGI performed with both systems, raw agreement for PMs was 79%, p = 0.250 for disagreement. CONCLUSIONS: ECGI systems were useful to identify the origin of the VAs, and the results were reproducible regardless the cardiac source.

Electrocardiographic imaging (ECGI): What is the minimal number of leads needed to obtain a good spatial resolution?

AIMS: Assess the minimal number of ECGI leads needed to obtain a good spatial resolution. METHODS: We enrolled 20 patients that underwent ablation of premature ventricular or atrial contractions using Carto and ECGI with AMYCARD. We evaluated the agreement regarding the site of origin of the arrhythmia between the ECGI and Carto, the area and diameter of the earliest activation site obtained with the ECGI (EASa and EASd). Based on previous studies with pacemapping, we considered a good spatial resolution of the ECGI when the EASd measured on the isopotential map was less than 18 mm. In presence of agreement the ECGI was reprocessed: a) with half the number of electrode bands (8 leads per electrode band) and b) with 6 electrode bands. RESULTS: The initial map was obtained with 23 (22-23) electrode bands per patient, corresponding to 143 (130-170) leads. Agreement rate was 85%, the median EASa and EASd were: 0.7 (0.5-1.3) cm2 and 9 (8-13) mm. With half the number of electrode bands including 73 (60-79) leads, agreement rate was 80%, the EASa and EASd were: 2.1 (1.5-6.2) cm2 and 16 (14 -28) mm. With only six electrode bands using 38 (30-42) leads, agreement rate was 55%, EASa and EASd were: 4.0 (3.3-5.0) cm2 and 23 (21-25) mm. The number of leads was a predictor of agreement with a good spatial resolution, OR (95% CI) of 1.138 (1.050-1.234), p = .002. According to the ROC curve, the minimal number of leads was 74 (AUC 0.981; 95% CI: 0.949-1.00, p < .0001). CONCLUSION: Reducing the number of leads was associated with a lower agreement rate and a significant reduction of spatial resolution. However, the number of leads needed to achieve a good spatial resolution was less than the maximal available.

Combination of lead-field theory with cardiac vector direction: ECG imaging of septal ventricular activation.

BACKGROUND: Despite the tremendous progress recently reported in ECG imaging (ECGI), some fundamental challenges are still hindering this non-invasive technology from meeting rising clinical expectations. In the present work, we address one of the major ECGI shortcomings in reconstruction of ventricular activation - the limited accuracy of endocardial and particularly septal mapping. METHODS: Ten CRT patients (five female, median (min-max) age - 61 (27-78) years) with previously implanted CRT devices underwent ECGI with isolated right ventricular (RV) pacing. In eight patients, the RV pacemaker lead was placed in the middle septal area of the posterior RV wall. Two subjects had a pacing lead in the anteroseptal apical segment, two at septal RVOT, two at septal junction with posterior wall and six in anterolateral segments. Lead positions were exactly known from CT scans, making the respective paced ECG sequences ideal for validation of ECGI endocardial accuracy. Non-invasive mapping was performed for single RV paced beats using original parameters of the CRT device. For non-invasive estimation of the focal origins, we considered the lead-field based fastest route algorithm (FRA) and its combination with the cardiac vector fit (FRA-V). Furthermore, we extended the resulting combined map by incorporating cardiac activation direction (FRA-V-D) provided by the cardiac dipole. RESULTS: The median (min-max) localization errors were 14 mm (7-27), 9 mm (7-28) and 11 mm (8-24) for FRA, FRA-V and FRA-V-D, respectively. Notably, in all cases at least one of the considered ECGI methods was able to correctly localize the found excitation origin on the endocardium. CONCLUSIONS: This preliminary study investigates combination of the rule-based fastest route algorithm with cardiac vector fit and direction for non-invasive imaging of septal ventricular sources. The developed ECGI methodology delivers reasonable reconstruction accuracy with the 10 mm median localization error. These findings suggest potential use of ECGI for challenging clinical cases, where catheter access to the correct cardiac anatomical region plays a crucial role in the execution of the electrophysiological procedure.

Non-invasive electrocardiographic imaging in patients with idiopathic premature ventricular contractions from the right ventricular outflow tract: New insights into arrhythmia substrate.

AIMS: The aim of this study was to use non-invasive electrocardiographic imaging (ECGI) to study the electrophysiological properties of right ventricular outflow tract (RVOT) in patients with frequent premature ventricular contractions (PVCs) from the RVOT and in controls. METHODS: ECGI is a combined application of body surface electrocardiograms and computed tomography or magnetic resonance imaging data. Unipolar electrograms are reconstructed on the epicardial and endocardial surfaces. Activation time (AT) was defined as the time of maximal negative slope of the electrogram (EGM) during QRS, recovery time (RT) as the time of maximal positive slope of the EGM during T wave, Activation recovery interval (ARI) was defined as the difference between RT and AT. ARI dispersion (Δ ARI) and RT dispersion (Δ RT) were calculated as the difference between maximal and minimal ARI and RT respectively. We evaluated those parameters in patients with frequent PVCs from the RVOT, defined as >10.000 per 24 h, and in a control group. RESULTS: We studied 7 patients with frequent RVOT PVCs and 17 controls. Patients with PVCs from the RVOT had shorter median RT than controls, in the endocardium and in the epicardium, respectively 380 (239-397) vs 414 (372-448) ms, p = 0.047 and 275 (236-301) vs 330 (263-418) ms, p = 0.047. The dispersion of ARI and of RT in the epicardium was higher than in controls, Δ ARI of 145 (68-216) vs 17 (3-48) ms, p = 0.001 and Δ RT of 201 (160-235) vs 115 (65-177), p = 0.019. CONCLUSION: In this group of patients we found a shorter median RT in the endocardium and in the epicardium of the RVOT and a higher dispersion of the ARI and RT across the epicardium in patients with PVCs from the RVOT when comparing to controls.

Fusion Imaging of Non-Invasive and Invasive Cardiac Electroanatomic Mapping in Patients with Ventricular Ectopic Beats: A Feasibility Analysis in a Case Series.

In patients with premature ventricular contractions (PVCs), non-invasive mapping could locate the PVCs' origin on a personalized 3-dimensional (3D) heart model and, thus, facilitate catheter ablation therapy planning. The aim of our report is to evaluate its accuracy compared to invasive mapping in terms of assessing the PVCs' early activation zone (EAZ). For this purpose, non-invasive electrocardiographic imaging (ECGI) was performed using the Amycard 01C system (EP Solutions SA, Switzerland) in three cases. In the first step, a multichannel ECG (up to 224 electrodes) was recorded, and the dominant PVCs were registered. Afterward, a cardiac computed tomography (in two cases) or magnetic resonance imaging (in one case) investigation was carried out acquiring non-contrast torso scans for 8-electrode strip visualization and contrast heart acquisition. For the reconstructed epi/endocardial meshes of the heart, non-invasive isochronal maps were generated for the selected multichannel ECG fragments. Then, the patients underwent an invasive electrophysiological study, and the PVCs' activation was evaluated by a 3D mapping system (EnSite NavX Precision, Abbott). Finally, using custom-written software, we performed 3D fusion of the non-invasive and invasive models and compared the resulting isochronal maps. A qualitative analysis in each case showed the same early localization of the dominant PVC on the endocardial surface when comparing the non-invasive and invasive isochronal maps. The distance from the EAZ to the mitral or tricuspid annulus was comparable in the invasive/non-invasive data (36/41 mm in case N1, 73/75 mm in case N2, 9/12 mm in case N3). The area of EAZ was also similar between the invasive/non-invasive maps (4.3/4.5 cm2 in case N1, 7.1/7.0 cm2 in case N2, 0.4/0.6 cm2 in case N3). The distances from the non-invasive to invasive earliest activation site were 4 mm in case N1, 7 mm in case N2, and 4 mm in case N3. Such results were appropriate to trust the clinical value of the preoperative data in these cases. In conclusion, the non-invasive identification of PVCs before an invasive electrophysiological study can guide clinical and interventional decisions, demonstrating appropriate accuracy in the estimation of focus origin.

Combination of personalized computational modeling and machine learning for optimization of left ventricular pacing site in cardiac resynchronization therapy.

Introduction: The 30-50% non-response rate to cardiac resynchronization therapy (CRT) calls for improved patient selection and optimized pacing lead placement. The study aimed to develop a novel technique using patient-specific cardiac models and machine learning (ML) to predict an optimal left ventricular (LV) pacing site (ML-PS) that maximizes the likelihood of LV ejection fraction (LVEF) improvement in a given CRT candidate. To validate the approach, we evaluated whether the distance DPS between the clinical LV pacing site (ref-PS) and ML-PS is associated with improved response rate and magnitude. Materials and methods: We reviewed retrospective data for 57 CRT recipients. A positive response was defined as a more than 10% LVEF improvement. Personalized models of ventricular activation and ECG were created from MRI and CT images. The characteristics of ventricular activation during intrinsic rhythm and biventricular (BiV) pacing with ref-PS were derived from the models and used in combination with clinical data to train supervised ML classifiers. The best logistic regression model classified CRT responders with a high accuracy of 0.77 (ROC AUC = 0.84). The LR classifier, model simulations and Bayesian optimization with Gaussian process regression were combined to identify an optimal ML-PS that maximizes the ML-score of CRT response over the LV surface in each patient. Results: The optimal ML-PS improved the ML-score by 17 ± 14% over the ref-PS. Twenty percent of the non-responders were reclassified as positive at ML-PS. Selection of positive patients with a max ML-score >0.5 demonstrated an improved clinical response rate. The distance DPS was shorter in the responders. The max ML-score and DPS were found to be strong predictors of CRT response (ROC AUC = 0.85). In the group with max ML-score > 0.5 and DPS< 30 mm, the response rate was 83% compared to 14% in the rest of the cohort. LVEF improvement in this group was higher than in the other patients (16 ± 8% vs. 7 ± 8%). Conclusion: A new technique combining clinical data, personalized heart modelling and supervised ML demonstrates the potential for use in clinical practice to assist in optimizing patient selection and predicting optimal LV pacing lead position in HF candidates for CRT.

Machine Learning Prediction of Cardiac Resynchronisation Therapy Response From Combination of Clinical and Model-Driven Data.

Background: Up to 30-50% of chronic heart failure patients who underwent cardiac resynchronization therapy (CRT) do not respond to the treatment. Therefore, patient stratification for CRT and optimization of CRT device settings remain a challenge. Objective: The main goal of our study is to develop a predictive model of CRT outcome using a combination of clinical data recorded in patients before CRT and simulations of the response to biventricular (BiV) pacing in personalized computational models of the cardiac electrophysiology. Materials and Methods: Retrospective data from 57 patients who underwent CRT device implantation was utilized. Positive response to CRT was defined by a 10% increase in the left ventricular ejection fraction in a year after implantation. For each patient, an anatomical model of the heart and torso was reconstructed from MRI and CT images and tailored to ECG recorded in the participant. The models were used to compute ventricular activation time, ECG duration and electrical dyssynchrony indices during intrinsic rhythm and BiV pacing from the sites of implanted leads. For building a predictive model of CRT response, we used clinical data recorded before CRT device implantation together with model-derived biomarkers of ventricular excitation in the left bundle branch block mode of activation and under BiV stimulation. Several Machine Learning (ML) classifiers and feature selection algorithms were tested on the hybrid dataset, and the quality of predictors was assessed using the area under receiver operating curve (ROC AUC). The classifiers on the hybrid data were compared with ML models built on clinical data only. Results: The best ML classifier utilizing a hybrid set of clinical and model-driven data demonstrated ROC AUC of 0.82, an accuracy of 0.82, sensitivity of 0.85, and specificity of 0.78, improving quality over that of ML predictors built on clinical data from much larger datasets by more than 0.1. Distance from the LV pacing site to the post-infarction zone and ventricular activation characteristics under BiV pacing were shown as the most relevant model-driven features for CRT response classification. Conclusion: Our results suggest that combination of clinical and model-driven data increases the accuracy of classification models for CRT outcomes.

His-Optimized Cardiac Resynchronization Therapy With Ventricular Fusion Pacing for Electrical Resynchronization in Heart Failure.

OBJECTIVES: This study sought to evaluate the effectiveness of His-optimized cardiac resynchronization therapy (HOT-CRT) for reducing left ventricular activation time (LVAT) compared to His bundle pacing (HBP) and biventricular (BiV) pacing (including multipoint pacing [MPP]), using electrocardiographic (ECG) imaging. BACKGROUND: HBP may correct bundle branch block (BBB) and has shown encouraging results for providing CRT. However, HBP does not correct BBB in all patients and may be combined with univentricular or BiV fusion pacing to deliver HOT-CRT to maximize resynchronization. METHODS: Nineteen patients with a standard indication for CRT, implanted with HBP without correction of BBB and BiV (n = 14) or right ventricular (n = 5) leads, were prospectively enrolled. Patients underwent ECG imaging while pacing in different configurations using different LV electrodes and at different HBP ventricular pacing (VP) delays. The primary endpoint was reduction in LVAT with HOT-CRT, and the secondary endpoints included various other dys-synchrony measurements including right ventricular activation time (RVAT). RESULTS: Compared to HBP, HOT-CRT reduced LVAT by 21% (-17 ms [95% confidence interval [CI]: -25 to -9 ms]; p < 0.001) and outperformed BiV by 24% (-22 ms [95% CI: -33 to -10 ms]; p = 0.002) and MPP by 13% (-11 ms [95% CI: -21 to -1 ms]; p = 0.035). Relative to HBP, HOT-CRT also reduced RVAT by 7% (-5 ms [95% CI: -9 to -1 ms; p = 0.035) in patients with right BBB, whereas RVAT was increased by BiV. The other electrical dyssynchrony measurements also improved with HOT-CRT. CONCLUSIONS: HOT-CRT acutely improves ventricular electrical synchrony beyond BiV and MPP. The impact of this finding needs to be evaluated further in studies with clinical follow-up. (Electrical Resynchronization and Acute Hemodynamic Effects of Direct His Bundle Pacing Compared to Biventricular Pacing; NCT03452462).

ECG Adapted Fastest Route Algorithm to Localize the Ectopic Excitation Origin in CRT Patients.

Although model-based solution strategies for the ECGI were reported to deliver promising clinical results, they strongly rely on some a priori assumptions, which do not hold true for many pathological cases. The fastest route algorithm (FRA) is a well-established method for noninvasive imaging of ectopic activities. It generates test activation sequences on the heart and compares the corresponding test body surface potential maps (BSPMs) to the measured ones. The test excitation propagation patterns are constructed under the assumption of a global conduction velocity in the heart, which is violated in the cardiac resynchronization (CRT) patients suffering from conduction disturbances. In the present work, we propose to apply dynamic time warping (DTW) to the test and measured ECGs before measuring their similarity. The warping step is a non-linear pattern matching that compensates for local delays in the temporal sequences, thus accounting for the inhomogeneous excitation propagation, while aligning them in an optimal way with respect to a distance function. To evaluate benefits of the temporal warping for FRA-based BSPMs, we considered three scenarios. In the first setting, a simplified simulation example was constructed to illustrate the temporal warping and display the resulting distance map. Then, we applied the proposed method to eight BSPMs produced by realistic ectopic activation sequences and compared its performance to FRA. Finally, we assessed localization accuracy of both techniques in ten CRT patients. For each patient, we noninvasively imaged two paced ECGs: from left and right ventricular implanted leads. In all scenarios, FRA-DTW outperformed FRA in terms of LEs. For the clinical cases, the median (25-75% range) distance errors were reduced from 16 (8-23)mm to 5 (2-10)mm for all pacings, from 15 (11-25)mm to 8 (3-13)mm in the left, and from 19 (6-23)mm to 4 (2-8)mm in the right ventricle, respectively. The obtained results suggest the ability of temporal ECG warping to compensate for an inhomogeneous conduction profile, while retaining computational efficiency intrinsic to FRA.